BIOL FSYCHIATRY 1992;31:135-146
135
Noradrenergic Function in Panic Dlsorc er Effects of Intravenous Clonidine Pretreatmen t on Lactate Induced Panic Jeremy D. Coplan, Michael R. Liebowitz, Jack M. Gorman, Abby J. Fyer, Donald J. Dillon, Raphael B. Campea~;, Sharon O. Davies, Jose Martinez, and Donald F. Klein
To assess the role of noradrenergic stimulation during lactate-induced panic, ten patients with panic disorder who panicked during a standard sodium-lactate ,i~fusion underwent a repeat infusion following intravenous clonidine pretreatment. Althc~gh clonidine significantly lowered prelactate systolic blood pressure, the drug did not ~'ignificantly lower prelactate anxiety levels, as reflected by the Acute Panic Inventor)' (API). Clonidine blocked lactate-induced panic in four of ten subjects, a significant effect. Clonidine treatment also significantly attenuated lactate-panic symptoms, as re~!lected by time to panic and AP! comparison between trials. Nevertheless, over half the ~ubjects still panicked in response to lactate despite clonidine. This preliminary study suggests that reduction of central noradrenergic activity by clonidine, at least at ~he dosage levels employed in the current study, only partially attenuates panic responses' to lactate. Noradrenergic theories of panic may not therefore fully account for lactaw panicogenesis.
Introduction Central noradrenergic dysregulation has been suggested as an underlying mechanism for panic disorder (PD) (Charney and Heninger 1986). Support for the "noradrenergic hypothesis" derives from observations that PD patients develop more frequent panic attacks than normal subjects following oral yohimbine ingestion (Chamey et al 1984). Yohimbine, an a2 antagonist (Svensson and Usdin 1978), provokes firing of the pontine nucleus locus ceruleus (LC) (Redmond 1977, 1979, Redmond and Huang 1979), a site containing the majority of central noradrenergic cell bodies. In accordance with this hypothesis, yohimbine-induced anxiety has been positively correlated with serum increases of the principal noradrenergic metabolite, 3-methyoxy 4-hydroxy phenylglycol (MHPG) (Charney and Heninger 1986; Gurguis and Uhde 1990). Conversely, acute clonidine administration is anxiolytic and antipanic (Charney and
From the New York State Psychiatric Institute and the Department of Psychiatry, College of Physicians and Surgeons, Columbia University, New York. Address reprint requests to Dr. Jeremy D. Coplan, Department of Psychiatry, Box 13, Columbia University, 722 West 168th Street, New York, NY 10032. Received March 11, 1991; revised July 11, 1991. © 1992 Society of Biological Psychiatry
0006-3223/92/$05.00
136
BIOL PSYCHIATRY 1992;31: ! 35-146
J.D. Coplan et al
Heninger 1986; Nutt 1989; Uhde et al 1989). Clonidine, an ot2 agonist, curtails central noradrenergic activity by reducing firing of LC neurons. The anxiolytic effect of clonidine is generally transient, as chronic oral clonidine administration is ineffective in a majority of PD patients (Uhde et al 1989; Liebowitz et al 1981). Tolerance to the anxiolytic effect of chronic clonidine treatment has been postulated to be due to desensitization of LC or2 autoreceptors t.Hsiao and Potter 1990). Despite the above evidence for a primary noradrenergic abnormality in panic disorder, several findings have not supported this theory. For instance, lactate-induced panic (Carr et al 1986), panic induced by carbon-dioxide inhalation (Woods et ai 1988), and panic induced by exposure to phobic stimuli (Woods et al 1987) all fail to produce consistent rises in serum MHPG, suggesting that increased LC activity is not regularly associated with these forms of panic. Furthermore, effective antipanic treatment, such as chronic imipramine (Klein and Fink 1962; Klein 1964), fails to block the anxiogenic effects of yohimbine (Charney et al 1985), and may even increase yohimbine anxiogenesis (Garfield et al 1967). Diazepam, on the other hand, an agent with limited antipanic properties (Noyes et al 1984), blocks yohimbine anxiogenesis in heakhy subjects (Charney et al 1984). Thus, the influence of pharmacological interventions on yohimbine response does not parallel response patterns observed in clinical treatment settings. Treatment studies therefore also suggest limitations of noradrenergic models of panic. Sodium lactate infusion reliably provokes panic in PD patients by unknown mechanisms (Pitts and McClure 1967, Liebowitz et al 1984). In the present study we wished to see if noradrenergic stimulation is a factor during lactate-induced panic. To test this hypothesis, PD patients who panicked during a sodium-lactate infusion were rechallenged with lactate following pretreatraent with intravenous clonidine. Methods
Subjects We studied ten subjects (Table 1), nine women and one man, aged 20-49 years, meeting DSM-III criteria for panic disorder or agoraphobia with panic attacks, who panicked during a standard sodium lactate infusion. Patients then returned for a second infusion procedure preceded by intravenous clonidine treatment. Duration of illness ranged from 3 months to 7 years, with a mean of 4.6 years. Exclusion and inclusion criteria were the same as for previous lactate studies (Liebowitz et al 1984). Written informed consent was obtained from each subject after full explanation of study procedures. Patients were informed that clonidine might lower blood pressure, but that its effects on lactate panic were not known. Patients were medication free for at least 2 weeks prior to the first infusion (except in the case of low-potency benzodiazepines, which were discontinued for at least 3 days prior to the first infusion).
Infusion Procedure The first infusion was given according to the previous standard lactate infusion protocol (Liebowitz et al 1984). Patients knew they would receive both a control substance and lactate, but not their respective timing. Staff in attendance were not blind to the infusion procedure.
Noradrenergic Function in Panic Disorder
BIOL PSYCHIATRY 1992;31:135-146
137
Table 1. Demographic Variables of Subjects" ID
Duration of Illness
192
Age
Sex
Wt
DX
CO-DX
7 yr
37
F
152
Ag/P
None
293 291
3 yr 1 yr
26 27
F F
100 127
Ag/P Ag/P
309
6 yr
39
F
136
Ag/P
319
2 yr
20
F
165
Ag/P
322
7 yr
35
F
140
Ag/P
14/O Dep. H/O Dep Secondary S.P. S.P. H/O Dep. Current Disthymia None
342
7 yr
26
F
141
Ag/P
363
3 mo
29
M
139
Ag/P
368
7 yr
49
F
141
Ag/P
356
6 yr
28
F
190
Ag/P
Secondary H/O Dep H/O marijuana abuse Irritable bowel syndrome H/O Dep. Secondary S.P.
Tx Behavioral therapy None Psychotherapy Prazepam Amitrypti. Oxazepam Psychotherapy Behavioral therapy Psychotherapy Oxazepam None Relaxation therapy Behavioral therapy
Last treatment >4 wk >4 wk 3 days
>4 wk >4 wk >4 wk >4 wk >4 wk >4 wk >4 wk
aS,p, -- social phobia, 14/O = Historyof, Dep -- Depression.
As per protocol, 28 min slow IV placebo infusion of normal saline was administered, then speeded up for 2 min to mimic subsequent lactate administration. At time 0, 0.5 mol/L racemic sodium lactate infusion was begun without the patient's knowledge. The total dosage of solution at 10 ml/kg of body weight was administered over 20 min unless the patients panicked, in which case the lactate administration was terminated (one patient who panicked at 22 min during a lactate infusion that extended beyond 20 min, because of a slow rate of administration, is included). Subjective symptoms were rated on a 28-item API scale, a modified and updated version of the API containing symptoms classically experienced during panic attacks (Dillon et al 1987). Each item was rated as not at all, mild, moderate, and severe and scored as 0,1,2, and 3, respectively. A maximum score of 84 was therefore obtainable on a given API. The criteria for panic were as described previously (Dillon et al 1987). The API was administered at both 35 and 5 min before lactate administration, at 10 and 20 min into the lactate infusion, or at the point of panic (POP), depending on the time to panic. Vital signs were monitored at - 3 0 , - 15, 0, 5, 10, 15, and 20 min and at the point of panic, with time 0 corresponding to the start of the lactate component of the infusion. Within 72 hr the subjects received a second standard sodium-lactate infusion. In this instance, however, the second infusion was preceded by clonidine pretreatment. In both infusions, patients were blind to the switch from normal saline to lactate, but patients were not blind to clonidine pretreatment for the second infusion. Similarly, response to the lactate-infusion following clonidine pretreatment was judged by a research psychiatrist not blind to clonidine pretreatment. Clonidine was diluted with saline and given by slow intravenous injection over a 10-
138
J.D. Coplan et al
BIOL PSYCHIATRY 1992,31:135-146
Table 2. Temporal Relationship of API Scores and Point o f Panic to Clonidine Dosing"
ID
192 293 291 309 319 322 342 363 368
Infusion Typeb
- 35
- 5
10
20
Pew
Time (min)
Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate. Clonidine Lactate Clonidine Lactate Cionidine
1 5 8 3 5 4 12 2 1 3 4 3 19 3 3 3 5 1
3 1 6 !2 13 8 15 6 1 8 8 16 32 7 5 11 3 7
27 13 18 20 18 15 56 39 14 19 34 23 55 29 25 16 35 29
-33 23 36 44 25 27 51 39 29 -22 ---35 -33
27 33 23 36 44 25 56 50 39 29 34 22 55 29 25 35 35 33
5 20 22 20 13 20 9 18 14 14 3 20 5 6 5 !9 5 13
Dose (ttg/kg) 2.0
1.0 1.0 1.0 1.5 1.5 1.5 1.5 1.5
°One subject's APi scores are missing. bLactate - first infusion of saline-lactate; clonidine -- second infusion of clonidine-lactate. ~Point of panic.
min period ending 45 min before the start of lactate. In contrast to the first saline infusion of 30 min, the second saline infusion lasted 45 min, in order to permit full effect of the clonidine to take place. Vital signs and the API were monitored at the same time points as the first infusion. The first patient received 2 I~g/kg clonidine and became quite hypotensive during the procedure. For this reason the next three patients were given 1 gg/kg clonidine, all without blood pressure difficulty. The remaining six received 1.5 ttg/kg clonidine and tolerated it well.
Data Analyses The two infusions permitted within-subject comparison of the response to saline (placebo) infusion with and without clonidine pretreatment and comparison of response to lactate with and without clonidine pretreatment (Table 2). Paired t tests were used for differences of prelactate infusion API and physiological measures between the untreated versus the clonidine treated saline infusions. The McNemar's test for correlated proportions was used for comparing the rate of panic between the first and second infusion. Paired t tests were used for duration of infusion between the two trial conditions. A 2 × 2 analysis of variance (ANOVA) for trial and time was used for the APl measures. Because one subject's API scores are missing, only nine subjects have been used for the API analyses. The first API analysis compared the two infusion conditions for the difference between prelactate API values and infusion termination API values (either at POP or end of infusion). The second API analysis compared the two infusion conditions for the difference between prelactate API values and API values 10 min into the lactate infusion. This
Noradrenergic Function in Panic Disorder
BIOL i~SYCHIATRY 1992;31:135-146
139
Table 3. Physiological Variablesa ID
291 309 319 322 342 356 363 368 Mean + SD Mean + SD
Infusion Type
Prelactate HR
Postlactate HR
Prelactate BP (mmHg)
Postlactate BP (mmHg)
Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate
74 68 92 61 57 50 56 52 53 50 61 68 51 54 67 58 63.88 - 13.65
93 89 103 92 m 90 62 111 58 83 78 84 83 61 96 82 81.86 - 17.08
114/70 105/48 129/70 104/75 119/69 85/53 115/73 95/57 122/71 101/65 113/72 111/71 111/44 101/65 123/80 106/60 118/68 -+ 6.1/10.5
121/60 1 | 3/67 183/108 i 71/90 128/64 104/53 131/73 108/68 133/96 111/67 139/79 120/79 124/61 135/72 142/68 i 12/58 135/76 _ 19.6/17.5
Clonidine
57.63 -- 7.44
86.50 -
101/62 -- 8.0/9.1
122/69 +- 22.0/11.6
13.83
~Two subject's physiological measures are missing.
analysis does not involve the second 10 min of either the first or second lactate infusion. The purpose of this analysis was to attempt to control for a possible imbalance of time to panic between the first and second infusion. If panic occurred before 10 min of the lactate infusion had elapsed, the API value at the POP was used for the 10-min value. The final APl analysis used paired t tests to compare the rate of lactate-induced increase of APl items between the two infusions. The rates values were determined by dividing the difference between end values and prelactate API values by duration of infusion. Although this final analysis has the shortcoming of assuming a linear slope for API increase, it does, in contrast to the ANOVA analyses, control for time effects more accurately. A 2 × 2 ANOVA for trial and time was used for physiological measures (systolic and diastolic blood pressure and heart rate). Data for physiological measures are only available on seven subjects (five of whom panicked and two who did not panic during the second infusion) (Table 3). The ANOVA compared physiological measures for both trials at 0 time and infusion termination (either at POP or end of infusion). Paired t tests were used to compare the rate of change (end value minus prelactate value divided by time) of physiological measures between trials. No Bonferroni correction was used. Significance levels reported are two-tailed.
Results Prelactate (Baseline) Values No significant differences for prelactate (baseline) APl values were observed between trials. Paired t tests indicate that clonidine treatment significantly lowered prelactate baseline systolic blood pressure (101.0 mmHg, SD = 7.9 versus 118.3 mmHg, SD =
140
J.D. Coplanet al
BIOL PSYCHIATRY 1992;31:135-146
SALINE PRETREATED LACTATE INFUSION VS. CLONIDINE PRETRF_.ATEDLACTATE INFUSION BASELINE TO TERMINATION OF INFUSION 3~
.,,0
i ........
34 D J 304u.l
~
O
~' 26 13. ,
135
Noradrenergic Function in Panic Dlsorc er Effects of Intravenous Clonidine Pretreatmen t on Lactate Induced Panic Jeremy D. Coplan, Michael R. Liebowitz, Jack M. Gorman, Abby J. Fyer, Donald J. Dillon, Raphael B. Campea~;, Sharon O. Davies, Jose Martinez, and Donald F. Klein
To assess the role of noradrenergic stimulation during lactate-induced panic, ten patients with panic disorder who panicked during a standard sodium-lactate ,i~fusion underwent a repeat infusion following intravenous clonidine pretreatment. Althc~gh clonidine significantly lowered prelactate systolic blood pressure, the drug did not ~'ignificantly lower prelactate anxiety levels, as reflected by the Acute Panic Inventor)' (API). Clonidine blocked lactate-induced panic in four of ten subjects, a significant effect. Clonidine treatment also significantly attenuated lactate-panic symptoms, as re~!lected by time to panic and AP! comparison between trials. Nevertheless, over half the ~ubjects still panicked in response to lactate despite clonidine. This preliminary study suggests that reduction of central noradrenergic activity by clonidine, at least at ~he dosage levels employed in the current study, only partially attenuates panic responses' to lactate. Noradrenergic theories of panic may not therefore fully account for lactaw panicogenesis.
Introduction Central noradrenergic dysregulation has been suggested as an underlying mechanism for panic disorder (PD) (Charney and Heninger 1986). Support for the "noradrenergic hypothesis" derives from observations that PD patients develop more frequent panic attacks than normal subjects following oral yohimbine ingestion (Chamey et al 1984). Yohimbine, an a2 antagonist (Svensson and Usdin 1978), provokes firing of the pontine nucleus locus ceruleus (LC) (Redmond 1977, 1979, Redmond and Huang 1979), a site containing the majority of central noradrenergic cell bodies. In accordance with this hypothesis, yohimbine-induced anxiety has been positively correlated with serum increases of the principal noradrenergic metabolite, 3-methyoxy 4-hydroxy phenylglycol (MHPG) (Charney and Heninger 1986; Gurguis and Uhde 1990). Conversely, acute clonidine administration is anxiolytic and antipanic (Charney and
From the New York State Psychiatric Institute and the Department of Psychiatry, College of Physicians and Surgeons, Columbia University, New York. Address reprint requests to Dr. Jeremy D. Coplan, Department of Psychiatry, Box 13, Columbia University, 722 West 168th Street, New York, NY 10032. Received March 11, 1991; revised July 11, 1991. © 1992 Society of Biological Psychiatry
0006-3223/92/$05.00
136
BIOL PSYCHIATRY 1992;31: ! 35-146
J.D. Coplan et al
Heninger 1986; Nutt 1989; Uhde et al 1989). Clonidine, an ot2 agonist, curtails central noradrenergic activity by reducing firing of LC neurons. The anxiolytic effect of clonidine is generally transient, as chronic oral clonidine administration is ineffective in a majority of PD patients (Uhde et al 1989; Liebowitz et al 1981). Tolerance to the anxiolytic effect of chronic clonidine treatment has been postulated to be due to desensitization of LC or2 autoreceptors t.Hsiao and Potter 1990). Despite the above evidence for a primary noradrenergic abnormality in panic disorder, several findings have not supported this theory. For instance, lactate-induced panic (Carr et al 1986), panic induced by carbon-dioxide inhalation (Woods et ai 1988), and panic induced by exposure to phobic stimuli (Woods et al 1987) all fail to produce consistent rises in serum MHPG, suggesting that increased LC activity is not regularly associated with these forms of panic. Furthermore, effective antipanic treatment, such as chronic imipramine (Klein and Fink 1962; Klein 1964), fails to block the anxiogenic effects of yohimbine (Charney et al 1985), and may even increase yohimbine anxiogenesis (Garfield et al 1967). Diazepam, on the other hand, an agent with limited antipanic properties (Noyes et al 1984), blocks yohimbine anxiogenesis in heakhy subjects (Charney et al 1984). Thus, the influence of pharmacological interventions on yohimbine response does not parallel response patterns observed in clinical treatment settings. Treatment studies therefore also suggest limitations of noradrenergic models of panic. Sodium lactate infusion reliably provokes panic in PD patients by unknown mechanisms (Pitts and McClure 1967, Liebowitz et al 1984). In the present study we wished to see if noradrenergic stimulation is a factor during lactate-induced panic. To test this hypothesis, PD patients who panicked during a sodium-lactate infusion were rechallenged with lactate following pretreatraent with intravenous clonidine. Methods
Subjects We studied ten subjects (Table 1), nine women and one man, aged 20-49 years, meeting DSM-III criteria for panic disorder or agoraphobia with panic attacks, who panicked during a standard sodium lactate infusion. Patients then returned for a second infusion procedure preceded by intravenous clonidine treatment. Duration of illness ranged from 3 months to 7 years, with a mean of 4.6 years. Exclusion and inclusion criteria were the same as for previous lactate studies (Liebowitz et al 1984). Written informed consent was obtained from each subject after full explanation of study procedures. Patients were informed that clonidine might lower blood pressure, but that its effects on lactate panic were not known. Patients were medication free for at least 2 weeks prior to the first infusion (except in the case of low-potency benzodiazepines, which were discontinued for at least 3 days prior to the first infusion).
Infusion Procedure The first infusion was given according to the previous standard lactate infusion protocol (Liebowitz et al 1984). Patients knew they would receive both a control substance and lactate, but not their respective timing. Staff in attendance were not blind to the infusion procedure.
Noradrenergic Function in Panic Disorder
BIOL PSYCHIATRY 1992;31:135-146
137
Table 1. Demographic Variables of Subjects" ID
Duration of Illness
192
Age
Sex
Wt
DX
CO-DX
7 yr
37
F
152
Ag/P
None
293 291
3 yr 1 yr
26 27
F F
100 127
Ag/P Ag/P
309
6 yr
39
F
136
Ag/P
319
2 yr
20
F
165
Ag/P
322
7 yr
35
F
140
Ag/P
14/O Dep. H/O Dep Secondary S.P. S.P. H/O Dep. Current Disthymia None
342
7 yr
26
F
141
Ag/P
363
3 mo
29
M
139
Ag/P
368
7 yr
49
F
141
Ag/P
356
6 yr
28
F
190
Ag/P
Secondary H/O Dep H/O marijuana abuse Irritable bowel syndrome H/O Dep. Secondary S.P.
Tx Behavioral therapy None Psychotherapy Prazepam Amitrypti. Oxazepam Psychotherapy Behavioral therapy Psychotherapy Oxazepam None Relaxation therapy Behavioral therapy
Last treatment >4 wk >4 wk 3 days
>4 wk >4 wk >4 wk >4 wk >4 wk >4 wk >4 wk
aS,p, -- social phobia, 14/O = Historyof, Dep -- Depression.
As per protocol, 28 min slow IV placebo infusion of normal saline was administered, then speeded up for 2 min to mimic subsequent lactate administration. At time 0, 0.5 mol/L racemic sodium lactate infusion was begun without the patient's knowledge. The total dosage of solution at 10 ml/kg of body weight was administered over 20 min unless the patients panicked, in which case the lactate administration was terminated (one patient who panicked at 22 min during a lactate infusion that extended beyond 20 min, because of a slow rate of administration, is included). Subjective symptoms were rated on a 28-item API scale, a modified and updated version of the API containing symptoms classically experienced during panic attacks (Dillon et al 1987). Each item was rated as not at all, mild, moderate, and severe and scored as 0,1,2, and 3, respectively. A maximum score of 84 was therefore obtainable on a given API. The criteria for panic were as described previously (Dillon et al 1987). The API was administered at both 35 and 5 min before lactate administration, at 10 and 20 min into the lactate infusion, or at the point of panic (POP), depending on the time to panic. Vital signs were monitored at - 3 0 , - 15, 0, 5, 10, 15, and 20 min and at the point of panic, with time 0 corresponding to the start of the lactate component of the infusion. Within 72 hr the subjects received a second standard sodium-lactate infusion. In this instance, however, the second infusion was preceded by clonidine pretreatment. In both infusions, patients were blind to the switch from normal saline to lactate, but patients were not blind to clonidine pretreatment for the second infusion. Similarly, response to the lactate-infusion following clonidine pretreatment was judged by a research psychiatrist not blind to clonidine pretreatment. Clonidine was diluted with saline and given by slow intravenous injection over a 10-
138
J.D. Coplan et al
BIOL PSYCHIATRY 1992,31:135-146
Table 2. Temporal Relationship of API Scores and Point o f Panic to Clonidine Dosing"
ID
192 293 291 309 319 322 342 363 368
Infusion Typeb
- 35
- 5
10
20
Pew
Time (min)
Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate. Clonidine Lactate Clonidine Lactate Cionidine
1 5 8 3 5 4 12 2 1 3 4 3 19 3 3 3 5 1
3 1 6 !2 13 8 15 6 1 8 8 16 32 7 5 11 3 7
27 13 18 20 18 15 56 39 14 19 34 23 55 29 25 16 35 29
-33 23 36 44 25 27 51 39 29 -22 ---35 -33
27 33 23 36 44 25 56 50 39 29 34 22 55 29 25 35 35 33
5 20 22 20 13 20 9 18 14 14 3 20 5 6 5 !9 5 13
Dose (ttg/kg) 2.0
1.0 1.0 1.0 1.5 1.5 1.5 1.5 1.5
°One subject's APi scores are missing. bLactate - first infusion of saline-lactate; clonidine -- second infusion of clonidine-lactate. ~Point of panic.
min period ending 45 min before the start of lactate. In contrast to the first saline infusion of 30 min, the second saline infusion lasted 45 min, in order to permit full effect of the clonidine to take place. Vital signs and the API were monitored at the same time points as the first infusion. The first patient received 2 I~g/kg clonidine and became quite hypotensive during the procedure. For this reason the next three patients were given 1 gg/kg clonidine, all without blood pressure difficulty. The remaining six received 1.5 ttg/kg clonidine and tolerated it well.
Data Analyses The two infusions permitted within-subject comparison of the response to saline (placebo) infusion with and without clonidine pretreatment and comparison of response to lactate with and without clonidine pretreatment (Table 2). Paired t tests were used for differences of prelactate infusion API and physiological measures between the untreated versus the clonidine treated saline infusions. The McNemar's test for correlated proportions was used for comparing the rate of panic between the first and second infusion. Paired t tests were used for duration of infusion between the two trial conditions. A 2 × 2 analysis of variance (ANOVA) for trial and time was used for the APl measures. Because one subject's API scores are missing, only nine subjects have been used for the API analyses. The first API analysis compared the two infusion conditions for the difference between prelactate API values and infusion termination API values (either at POP or end of infusion). The second API analysis compared the two infusion conditions for the difference between prelactate API values and API values 10 min into the lactate infusion. This
Noradrenergic Function in Panic Disorder
BIOL i~SYCHIATRY 1992;31:135-146
139
Table 3. Physiological Variablesa ID
291 309 319 322 342 356 363 368 Mean + SD Mean + SD
Infusion Type
Prelactate HR
Postlactate HR
Prelactate BP (mmHg)
Postlactate BP (mmHg)
Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate Clonidine Lactate
74 68 92 61 57 50 56 52 53 50 61 68 51 54 67 58 63.88 - 13.65
93 89 103 92 m 90 62 111 58 83 78 84 83 61 96 82 81.86 - 17.08
114/70 105/48 129/70 104/75 119/69 85/53 115/73 95/57 122/71 101/65 113/72 111/71 111/44 101/65 123/80 106/60 118/68 -+ 6.1/10.5
121/60 1 | 3/67 183/108 i 71/90 128/64 104/53 131/73 108/68 133/96 111/67 139/79 120/79 124/61 135/72 142/68 i 12/58 135/76 _ 19.6/17.5
Clonidine
57.63 -- 7.44
86.50 -
101/62 -- 8.0/9.1
122/69 +- 22.0/11.6
13.83
~Two subject's physiological measures are missing.
analysis does not involve the second 10 min of either the first or second lactate infusion. The purpose of this analysis was to attempt to control for a possible imbalance of time to panic between the first and second infusion. If panic occurred before 10 min of the lactate infusion had elapsed, the API value at the POP was used for the 10-min value. The final APl analysis used paired t tests to compare the rate of lactate-induced increase of APl items between the two infusions. The rates values were determined by dividing the difference between end values and prelactate API values by duration of infusion. Although this final analysis has the shortcoming of assuming a linear slope for API increase, it does, in contrast to the ANOVA analyses, control for time effects more accurately. A 2 × 2 ANOVA for trial and time was used for physiological measures (systolic and diastolic blood pressure and heart rate). Data for physiological measures are only available on seven subjects (five of whom panicked and two who did not panic during the second infusion) (Table 3). The ANOVA compared physiological measures for both trials at 0 time and infusion termination (either at POP or end of infusion). Paired t tests were used to compare the rate of change (end value minus prelactate value divided by time) of physiological measures between trials. No Bonferroni correction was used. Significance levels reported are two-tailed.
Results Prelactate (Baseline) Values No significant differences for prelactate (baseline) APl values were observed between trials. Paired t tests indicate that clonidine treatment significantly lowered prelactate baseline systolic blood pressure (101.0 mmHg, SD = 7.9 versus 118.3 mmHg, SD =
140
J.D. Coplanet al
BIOL PSYCHIATRY 1992;31:135-146
SALINE PRETREATED LACTATE INFUSION VS. CLONIDINE PRETRF_.ATEDLACTATE INFUSION BASELINE TO TERMINATION OF INFUSION 3~
.,,0
i ........
34 D J 304u.l
~
O
~' 26 13. ,
