BIOL PSYCHIATRY 1990;28:315-324
315
Effect of the Anesthetic Agent Propofol on Hormonal Responses to ECT Philip Mitchell, George Smythe, and Thomas Torda
Propefol is a new anesthetic induction agent ttmt red,{ces electroconvulsive therapy (ECT) seizure duration. To indirectly investigate the effect of propofol on ECT-induced acute central neurotransmitter changes, we studied neuroendocrine responses in 25 primary depressed subjects treated with ECT under either propofol or thiopentone anesthesia. Blood samples were taken prior to ECT, and then at regular intervals for 2 hr. Only the prolactin response correlated significantly with seizure duration (r = 0.52, p < 0.01). Subjects given pvopofol had significantly reduced adrenocorticotropin (ACTH) (p < 0.(71) and cortisol (p < 0.05) responses compared to thiopentone, which were independent of seizure duration. There was a trend towards a reduction in the prolactin response wt~h propofol compared to thiopent~ne, but this ~'-~ der.cndent upon the diminished seizure duration. The results ;.ndicate that propofol affects endocrine responses to ECT by two distinct mechanisms: decreasing prolactin by reducing the seizure duration and decreasing ACTH and cortisol by another process, possibly via a reduction in central noradrenergic activation.
Introduction Propofol is an intravenous anesthetic agent that has recently been introduced into clinical practice. The main advan'~ages relevant to electroeonvulsive therapy (ECT) use are the extremely rapid recovery (Crankshaw 1988) and reduced post-ECT hypertension compared to the short-acting barbiturates (e.g., Bone et al. 1988; Rampton et al. 1989). Recent reports have indicated, however, that propofoi reduces ECT seizure duration (Dwyer et al. 1988; Rampton et al. 1989; Rouse 1988; Simpson et al. 1988), leading to concern (Simpson and Snaith 1989) that propofol may reduce the efficacy of ECT and that it should therefore be contraindicated. In a recent study (Mitchell et al., unpublished data) we examined the effect of propofol and thio~ntone on the length of course of ECT and the efficacy of this treatment. Though confirming the reduced seizure duration demonstrated in previous studies, we found no evidence of reduced efficacy as measured by percentage or absolute reduction in Hamilton
From the School of Psychiatry, University of New South Wales, Kensington (P.M.): the Garvan Institute of Medical Research, St. Vincent's Hospital, Darlinghurst (G.S.)~ aad the Department of Anaesthesia and Intensive Care, Prince Henry Hospital. Little Bay, N.S.W., Australia (T.T.). Supported in part by the Australian National Health and Medical Research Council. Address reprint requests to Dr. P. B. Mitchell, Mood Disorders Unit, Prince Henry Hospital, Anzac Parade. Little Bay, N.S.W., 2036, Australia. Received November 18, 1989; revised February 12, 1990.
© 1990 Society of Biological Psychiatry
0006-3223/90/$03.50
316
BIOLPSYCHIATRY
P. Mitchell et al.
1990;28:315-324
and Zung scores over the ECT course. The duration of course of ECT was prolonged with propofol anesthesia, but this difference was not significant. Hormonal responses have been used as an indirect index of acute ECT-induced changes in central neurochemicai function with increased levels of prolactin, cortisol, adrenocorticotropin (ACTH), vasopressin, and estrogen-stimulated neurophysin (see Deakin et al. 1983; Whalley et al. 1987; Weizman et al. 1987) being described. Controlled trials have confirmed that these responses are independent of the normal endocrine responses to anesthesia (e.g., Deakin et al. 1983). There have been variable findings for growth hormone, with some studies demonstrating a reduction in levels, and others showing no change. The specific neurotransmitter changes underlying these hormonal responses have not been determined, although Whalley et al. (1987) have suggested that they are consistent with central serotonergic activation. The aim of this study was to compare the effects of propofol and the short-acting barbiturate thiopentone on neuroendocrine responses to ECT by examining data from a prospective ECT study undertaken prior to our awareness of the propofol ECT reports. Methods
Subjects The sample comprised 17 women and 8 men with a mean age ( _ SD) of 58.5 _+ 14.7 years. These 25 subjects were a subgroup of the 70 reported in our previous study (Mitchell et al. unpublished data). Diagnoses were made following a standardized semistructured interview for depression (Brodaty et al. 1987), and all patients satisfied DSM-III criteria for major depression.
Anesthetic and ECT Details The study period extended over 30 months, being 15 months prior and 15 months subsequent to the introduction of propofol in our hospital in March 1988. The anesthetic component of this study was "naturalistic" in that choice of anesthetic was determined by the individual preference of the anesthetist and not by the diagnosis, age, or physical status of the patient. No premedication was given and the patients were preoxygenated for 2 min. Fifteen patients had ECT under thiopentone sodium and 10 under propofol anesthesia. Both induction agents were administered by slow intravenous injection until the eye-lash reflex was lost. Suxamethonium, approximately 0.5 mg/kg, was used as a muscle relaxant in all subjects. Before induction of the seizure, several breaths of oxygen were administered from a Laerdahl resuscitator, and positive pressure ventilation at approximately 20 breaths per min was maintained until spontaneous ventilation resumed. S,'.!zures were induced using a Kabtronics Model 271e ECT apparatus, which delivers fixed current (800 mA) square-wave pulses of 0.5 msec duration at a rate of 125/sec for 3 sec. Unilateral electrode placement over the nondominant hemisphere was used. Seizure duration was monitored clinically by an experienced psychiatric nurse without use of the "cuff" (isolated arm) method (Adderley and Hamilton 1953, Fink and Johnson 1982). Only bilateral seizures were considered a6equate and durations of both the tonic and clonic phases (after cessation of the electrical stimulus) were recorded. These measurements were made prior to the unit's awareness of the findings of reduced seizure duration with propofol.
Hormonal Response to ECT: Effect of Propofol
BIOL PSYCHIATRY 1990;27:315-324
317
T a b l e I. D e t a i l s o f A n e s t h e t i c G r o u p s
Age (years 4- SD) Gender (male/female) Major depression with melancholia Suxamethonium dosage (mg 4- SD) Induction agent dosage (mg/kg 4- SD)
Thiopentone (n = 15)
Pmpofol (n = 10)
62.3 ---- 13.3 6/9 14 34.7 4- 9.0 2.88 "4" 0.66
52.9 -- 15.6" 2/8 b 10~ 38.5 ---- 5.3 a 1.5! "4- 0.34 e
at = 1.61, p = 0.12; bX2 = 0.38, p = 0.54; cX2 = 0.00, p = !.00; dl = -1.21, p = 0.24; ethiopentone/propofol dosage ratio = 1.91.
Endocrine Measurements An indwelling cannula was inserted into an antecubital vein 1 hr prior to the first ECT. Subjects were only studied on this one cessation. Baseline blood samples were collected 30 min before and immediately prior to injection of the anesthetic agent. Post-ECT samples were then collected at 2, 4, 6, 8, 10, 15, 30, 60 and 120 min, with the exception of samples for ACTH which were collected at 6, 10, 15, and 30 min. Blood was centrifuged at 4°C for 10 min at 1000 g within 1 hr of collection and the serum was stored at - 20°C. Serum prolactin levels were measured using a double antibody radioimmunoassay with an interassay coefficient of variation (cv) of 5% and intraassay cv of 6.5% at 20.9 ng/ml. Serum cortisol levels were measured by radioimmunoassay of the total cortisol, with an intraassay cv of 6.8% and an interassay cv of 7.9% at 13.6 ~g/dl. Plasma ACTH levels were measured using a radioimmunoassay, with an intraassay cv of 13.2% at 26.6 ng/liter and an interassay cv of 23.5% at 25.4 rig/liter (only one of the 25 subjects did not have plasma assayed for ACTH). Growth hormone levels were measured using polyclonal antiserum with an intraassay cv of 5.8% and an interassay cv of 10.9% at 1.98 ng/ml.
Statistical Methods Endocrine responses were analyzed as the absolute increase to peak above baseline (A) for all hormones except growth hormone for which the absolute reduction (-~ A) was used. The frequency distributions of the examined variables were assessed by means of the Kolmogorov-Smirnov test. All data were normally distributed apart from the ACTH and growth hormone responses. For normally distributed data, Student's t-test and Pearson's correlation coefficient were used. For nonnormally distributed data, variables were loglo-transfo~nned prior to use of t-tests, and Spearman's rank order correlations (rs) were used to assess interrelationships between untransformed variables. Analysis of covariance (ANCOVA) was used to determine the effect of seizure duration on hormone responses. All endocrine responses were log~o-transformed prior to ANCOVA. Results
Anesthetic and Clinical Details The average anesthetic dosages (-+ SD) were as follows: thiopentone 171.0 - 36.5 mg, propofol 94.0 -+ 32.0 mg, and suxamethonium 36.2 - 7.8 mg. Details of the two anesthetic groups are outlined in Table 1. There were no significant differences between the groups with regard to age, gender, DSM-III diagnosis, or suxamethonium dosage.
318
P. Mitchell et al.
BIOL PSYCHIATRY 1990:28:315-324
Table 2. Endocrine Responses to ECT Significance
Anesthetic agent
Hormone
A A C T H (ng/liter)
Thiopentone (n = 15) With outlier" 128.7 +_ 231.6 Without outlier b 69.2 +_ 65.4
A CortisoJ (?.g/dl) A Prolactin (ng/ml) - A Growth hormone (ng/ml)
8.7 -
3.7
30.1 ± 25.8 -1.1
-
2.6
Propofoi (n = 10)
12.2 -
t-test
17.0
12.2 _ 17.0 5.0 _ 4.7 16.9-2.5
ANCOVA (with seizure duration as covariate)
16.6 ± 3.5
t = 3.46 p < 0.01
F = 22.86
t p t p t p t p
F = 19.40 p < 0.001 F = 6.81 p < 0.05 F=0.60 p = 0.45 F = 0.47 p = 0.50
ffi < = < = = = =
3.29 0.01 2.22 0.05 1.52 0.14 1.26 0.22
p < 0.001
Raw means _ SD. °Data logm-transformed prior to statistical analyses; n = 14 (thiopentone); 10 (propofol). bn = 13 (thiopentone); 10 (propofol). CData Iog~0-transformed prior to statistical analyses.
Dosages in mg/kg for each anesthetic induction agent are also outlined in Table 1. The ratio of the mean dose~ of the two induction agents actually administered, 1.91, is very close to the published equipotency ratio for bolus doses of 1.61 (Grounds et al. 1986).
Seizure Duration The mean duration ( _+SD) of seizures after propofol was 20.0 _+ 11.5 sec compared to 28.6 +_ 9.6 sec after thiopentone (t = 2.02, p = 0.06). No patients failed to have a seizure.
ACTH/Cortisol There were no significant correlations between seizure duration and increase in either ACTH (rs - - 0 . 0 2 ) or cortisol (r = 0.18). ACTH increased significantly with both thiopentone (t = - 3 . 8 1 , p < 0.01) and propofol (t -- - 2 . 3 7 , p < 0.05), as did cortisol (t = - 9 . 1 0 , p < 0.001; and t = - 3 . 3 7 , p < 0.01, respectively). As indicated in Table 2 and Figures 1 and 2, however, these responses were significantly attenuated with propofol anesthesia compared to thiopentone. The reduction in ACTH response with propofol remained significant even when one outlier from the thiopentone group, who had an increase of 900 ng/liter after ECT, was excluded. Analysis of covariance, with seizure duration as a covariate, indicated that the reduction in ACTH was independent of the propofi~!-induced shortened seizure duration (p < 0.001). Similarly, the cortisol response was s~ill significant after covarying for seizure duration (p < 0.05). ANCOVAs with either age or suxamethonium dosage as covariates also demonstrated that the difference in ACTH responses between the two groups was not related to these factors (F = 8.32, df -- 1,21, p < 0.01; F - 11.03, df - 1,21 p < 0.01).
Hormonal Response to ECT: Effect of Propofol
150
120
BIOL PSYCHIATRY 1990;27:3i5-324
319
-
-
\
oD
I= =c I-0
90
-
315
Effect of the Anesthetic Agent Propofol on Hormonal Responses to ECT Philip Mitchell, George Smythe, and Thomas Torda
Propefol is a new anesthetic induction agent ttmt red,{ces electroconvulsive therapy (ECT) seizure duration. To indirectly investigate the effect of propofol on ECT-induced acute central neurotransmitter changes, we studied neuroendocrine responses in 25 primary depressed subjects treated with ECT under either propofol or thiopentone anesthesia. Blood samples were taken prior to ECT, and then at regular intervals for 2 hr. Only the prolactin response correlated significantly with seizure duration (r = 0.52, p < 0.01). Subjects given pvopofol had significantly reduced adrenocorticotropin (ACTH) (p < 0.(71) and cortisol (p < 0.05) responses compared to thiopentone, which were independent of seizure duration. There was a trend towards a reduction in the prolactin response wt~h propofol compared to thiopent~ne, but this ~'-~ der.cndent upon the diminished seizure duration. The results ;.ndicate that propofol affects endocrine responses to ECT by two distinct mechanisms: decreasing prolactin by reducing the seizure duration and decreasing ACTH and cortisol by another process, possibly via a reduction in central noradrenergic activation.
Introduction Propofol is an intravenous anesthetic agent that has recently been introduced into clinical practice. The main advan'~ages relevant to electroeonvulsive therapy (ECT) use are the extremely rapid recovery (Crankshaw 1988) and reduced post-ECT hypertension compared to the short-acting barbiturates (e.g., Bone et al. 1988; Rampton et al. 1989). Recent reports have indicated, however, that propofoi reduces ECT seizure duration (Dwyer et al. 1988; Rampton et al. 1989; Rouse 1988; Simpson et al. 1988), leading to concern (Simpson and Snaith 1989) that propofol may reduce the efficacy of ECT and that it should therefore be contraindicated. In a recent study (Mitchell et al., unpublished data) we examined the effect of propofol and thio~ntone on the length of course of ECT and the efficacy of this treatment. Though confirming the reduced seizure duration demonstrated in previous studies, we found no evidence of reduced efficacy as measured by percentage or absolute reduction in Hamilton
From the School of Psychiatry, University of New South Wales, Kensington (P.M.): the Garvan Institute of Medical Research, St. Vincent's Hospital, Darlinghurst (G.S.)~ aad the Department of Anaesthesia and Intensive Care, Prince Henry Hospital. Little Bay, N.S.W., Australia (T.T.). Supported in part by the Australian National Health and Medical Research Council. Address reprint requests to Dr. P. B. Mitchell, Mood Disorders Unit, Prince Henry Hospital, Anzac Parade. Little Bay, N.S.W., 2036, Australia. Received November 18, 1989; revised February 12, 1990.
© 1990 Society of Biological Psychiatry
0006-3223/90/$03.50
316
BIOLPSYCHIATRY
P. Mitchell et al.
1990;28:315-324
and Zung scores over the ECT course. The duration of course of ECT was prolonged with propofol anesthesia, but this difference was not significant. Hormonal responses have been used as an indirect index of acute ECT-induced changes in central neurochemicai function with increased levels of prolactin, cortisol, adrenocorticotropin (ACTH), vasopressin, and estrogen-stimulated neurophysin (see Deakin et al. 1983; Whalley et al. 1987; Weizman et al. 1987) being described. Controlled trials have confirmed that these responses are independent of the normal endocrine responses to anesthesia (e.g., Deakin et al. 1983). There have been variable findings for growth hormone, with some studies demonstrating a reduction in levels, and others showing no change. The specific neurotransmitter changes underlying these hormonal responses have not been determined, although Whalley et al. (1987) have suggested that they are consistent with central serotonergic activation. The aim of this study was to compare the effects of propofol and the short-acting barbiturate thiopentone on neuroendocrine responses to ECT by examining data from a prospective ECT study undertaken prior to our awareness of the propofol ECT reports. Methods
Subjects The sample comprised 17 women and 8 men with a mean age ( _ SD) of 58.5 _+ 14.7 years. These 25 subjects were a subgroup of the 70 reported in our previous study (Mitchell et al. unpublished data). Diagnoses were made following a standardized semistructured interview for depression (Brodaty et al. 1987), and all patients satisfied DSM-III criteria for major depression.
Anesthetic and ECT Details The study period extended over 30 months, being 15 months prior and 15 months subsequent to the introduction of propofol in our hospital in March 1988. The anesthetic component of this study was "naturalistic" in that choice of anesthetic was determined by the individual preference of the anesthetist and not by the diagnosis, age, or physical status of the patient. No premedication was given and the patients were preoxygenated for 2 min. Fifteen patients had ECT under thiopentone sodium and 10 under propofol anesthesia. Both induction agents were administered by slow intravenous injection until the eye-lash reflex was lost. Suxamethonium, approximately 0.5 mg/kg, was used as a muscle relaxant in all subjects. Before induction of the seizure, several breaths of oxygen were administered from a Laerdahl resuscitator, and positive pressure ventilation at approximately 20 breaths per min was maintained until spontaneous ventilation resumed. S,'.!zures were induced using a Kabtronics Model 271e ECT apparatus, which delivers fixed current (800 mA) square-wave pulses of 0.5 msec duration at a rate of 125/sec for 3 sec. Unilateral electrode placement over the nondominant hemisphere was used. Seizure duration was monitored clinically by an experienced psychiatric nurse without use of the "cuff" (isolated arm) method (Adderley and Hamilton 1953, Fink and Johnson 1982). Only bilateral seizures were considered a6equate and durations of both the tonic and clonic phases (after cessation of the electrical stimulus) were recorded. These measurements were made prior to the unit's awareness of the findings of reduced seizure duration with propofol.
Hormonal Response to ECT: Effect of Propofol
BIOL PSYCHIATRY 1990;27:315-324
317
T a b l e I. D e t a i l s o f A n e s t h e t i c G r o u p s
Age (years 4- SD) Gender (male/female) Major depression with melancholia Suxamethonium dosage (mg 4- SD) Induction agent dosage (mg/kg 4- SD)
Thiopentone (n = 15)
Pmpofol (n = 10)
62.3 ---- 13.3 6/9 14 34.7 4- 9.0 2.88 "4" 0.66
52.9 -- 15.6" 2/8 b 10~ 38.5 ---- 5.3 a 1.5! "4- 0.34 e
at = 1.61, p = 0.12; bX2 = 0.38, p = 0.54; cX2 = 0.00, p = !.00; dl = -1.21, p = 0.24; ethiopentone/propofol dosage ratio = 1.91.
Endocrine Measurements An indwelling cannula was inserted into an antecubital vein 1 hr prior to the first ECT. Subjects were only studied on this one cessation. Baseline blood samples were collected 30 min before and immediately prior to injection of the anesthetic agent. Post-ECT samples were then collected at 2, 4, 6, 8, 10, 15, 30, 60 and 120 min, with the exception of samples for ACTH which were collected at 6, 10, 15, and 30 min. Blood was centrifuged at 4°C for 10 min at 1000 g within 1 hr of collection and the serum was stored at - 20°C. Serum prolactin levels were measured using a double antibody radioimmunoassay with an interassay coefficient of variation (cv) of 5% and intraassay cv of 6.5% at 20.9 ng/ml. Serum cortisol levels were measured by radioimmunoassay of the total cortisol, with an intraassay cv of 6.8% and an interassay cv of 7.9% at 13.6 ~g/dl. Plasma ACTH levels were measured using a radioimmunoassay, with an intraassay cv of 13.2% at 26.6 ng/liter and an interassay cv of 23.5% at 25.4 rig/liter (only one of the 25 subjects did not have plasma assayed for ACTH). Growth hormone levels were measured using polyclonal antiserum with an intraassay cv of 5.8% and an interassay cv of 10.9% at 1.98 ng/ml.
Statistical Methods Endocrine responses were analyzed as the absolute increase to peak above baseline (A) for all hormones except growth hormone for which the absolute reduction (-~ A) was used. The frequency distributions of the examined variables were assessed by means of the Kolmogorov-Smirnov test. All data were normally distributed apart from the ACTH and growth hormone responses. For normally distributed data, Student's t-test and Pearson's correlation coefficient were used. For nonnormally distributed data, variables were loglo-transfo~nned prior to use of t-tests, and Spearman's rank order correlations (rs) were used to assess interrelationships between untransformed variables. Analysis of covariance (ANCOVA) was used to determine the effect of seizure duration on hormone responses. All endocrine responses were log~o-transformed prior to ANCOVA. Results
Anesthetic and Clinical Details The average anesthetic dosages (-+ SD) were as follows: thiopentone 171.0 - 36.5 mg, propofol 94.0 -+ 32.0 mg, and suxamethonium 36.2 - 7.8 mg. Details of the two anesthetic groups are outlined in Table 1. There were no significant differences between the groups with regard to age, gender, DSM-III diagnosis, or suxamethonium dosage.
318
P. Mitchell et al.
BIOL PSYCHIATRY 1990:28:315-324
Table 2. Endocrine Responses to ECT Significance
Anesthetic agent
Hormone
A A C T H (ng/liter)
Thiopentone (n = 15) With outlier" 128.7 +_ 231.6 Without outlier b 69.2 +_ 65.4
A CortisoJ (?.g/dl) A Prolactin (ng/ml) - A Growth hormone (ng/ml)
8.7 -
3.7
30.1 ± 25.8 -1.1
-
2.6
Propofoi (n = 10)
12.2 -
t-test
17.0
12.2 _ 17.0 5.0 _ 4.7 16.9-2.5
ANCOVA (with seizure duration as covariate)
16.6 ± 3.5
t = 3.46 p < 0.01
F = 22.86
t p t p t p t p
F = 19.40 p < 0.001 F = 6.81 p < 0.05 F=0.60 p = 0.45 F = 0.47 p = 0.50
ffi < = < = = = =
3.29 0.01 2.22 0.05 1.52 0.14 1.26 0.22
p < 0.001
Raw means _ SD. °Data logm-transformed prior to statistical analyses; n = 14 (thiopentone); 10 (propofol). bn = 13 (thiopentone); 10 (propofol). CData Iog~0-transformed prior to statistical analyses.
Dosages in mg/kg for each anesthetic induction agent are also outlined in Table 1. The ratio of the mean dose~ of the two induction agents actually administered, 1.91, is very close to the published equipotency ratio for bolus doses of 1.61 (Grounds et al. 1986).
Seizure Duration The mean duration ( _+SD) of seizures after propofol was 20.0 _+ 11.5 sec compared to 28.6 +_ 9.6 sec after thiopentone (t = 2.02, p = 0.06). No patients failed to have a seizure.
ACTH/Cortisol There were no significant correlations between seizure duration and increase in either ACTH (rs - - 0 . 0 2 ) or cortisol (r = 0.18). ACTH increased significantly with both thiopentone (t = - 3 . 8 1 , p < 0.01) and propofol (t -- - 2 . 3 7 , p < 0.05), as did cortisol (t = - 9 . 1 0 , p < 0.001; and t = - 3 . 3 7 , p < 0.01, respectively). As indicated in Table 2 and Figures 1 and 2, however, these responses were significantly attenuated with propofol anesthesia compared to thiopentone. The reduction in ACTH response with propofol remained significant even when one outlier from the thiopentone group, who had an increase of 900 ng/liter after ECT, was excluded. Analysis of covariance, with seizure duration as a covariate, indicated that the reduction in ACTH was independent of the propofi~!-induced shortened seizure duration (p < 0.001). Similarly, the cortisol response was s~ill significant after covarying for seizure duration (p < 0.05). ANCOVAs with either age or suxamethonium dosage as covariates also demonstrated that the difference in ACTH responses between the two groups was not related to these factors (F = 8.32, df -- 1,21, p < 0.01; F - 11.03, df - 1,21 p < 0.01).
Hormonal Response to ECT: Effect of Propofol
150
120
BIOL PSYCHIATRY 1990;27:3i5-324
319
-
-
\
oD
I= =c I-0
90
-
