Effects of Buspirone on Sleep and Respiration1 , 2
WALLACE B. MENDELSON, JOSEPH V. MARTIN, and DAVID M. RAPOPORT
Introduction
Anxiolytic agents, such as the benzodiazepines, are among the most widely used drugs. In addition to concern about unwanted sedation in patients treated with many of these agents, it has been increasingly recognized that anxiolytics depress ventilation, particularly in patients with sleep-related respiratory impairment or chronic obstructive pulmonary disease (1). The azospirone buspirone is a nonbenzodiazepine that possesses anticonflict and antiaggressive effects in animals (2) and is anxiolytic in humans (3). Clinical reports have indicated that it is nonsedating (3, 4), and a polygraphic study found no significant effects of clinically used doses on the sleep of insomniacs (5), except to delay the onset and reduce the amount of rapid eye movement (REM) sleep. In rats, somewhat higher doses may actually decrease sleep (6). No effect of buspirone on the appearance of electroencephalographic (EEG) waveforms inside individual states of consciousness (waking, non-REM, and REM) has been reported. A preliminary clinical study has indicated that buspirone does not suppress ventilation (7). More recently,Garner and coworkers (8) have reported that buspirone stimulates respiration as measured by phrenic nerve responses in anesthetized, paralyzed cats whose carotid sinus nerves and vagi had been cut. To extend these observations to the intact and sleeping animal, we have examined the effects of two doses of buspirone on sleep and respiration in unanesthetized, freely moving rats. Method Sleep Recordings A group of 10male Sprague-Dawley rats (250 to 300g) wereanesthetized (70 mg/kg of ketamine and 6 mg/kg of xylazine, intramuscularly) and surgically implanted with 0-80 stainless steel screwsthat served as dural EEG electrodes; Teflons'-coated 0.010 inch stainless steel wires were embedded in the nuchal musculature for electromyographic recording (9). After a l-wk recovery period in chambers with lights on from 8:00 A.M. to 8:00 P.M.,
SUMMARY Drugs used in the treatment of anxiety are frequently sedating and tend to be respiratory depressants. Buspirone, a nonbenzodiazepine anxiolytic agent, has little reported sedative effect. It has been shown to be a respiratory stimulant in an anesthetized, glomectomized cat model. In this study, we examined the effects of two intraperitoneal single doses (10and 20 mg/kg) of buspirone on sleep and respiration in unanesthetized, intact, freely moving rats. Buspirone increased sleep latency (p < 0.0001)and decreased total sleep (p < 0.02) through reductions in both non-REM and REM sleep. Respiratory rate (p < 0.0003) and ventilation (p < 0.004) were significantly increased for 4 h after drug injection. The effects on respiration were independent of those on sleep; stimulation was evident in both waking and non-REM sleep. This study suggests that buspirone, in addition to being free of sedating and respiratory depressant side effects when prescribed for anxiety in humans, may be a respiratory stimulant whose effects persist in sleep. AM REV RESPIR DIS 1990; 141:1527-1530
animals were injected with vehicle (0.90/0 saline) or intraperitoneal buspirone, 10 or 20 mg/kg, at 9:00 A.M., after which sleep and respiration were recorded for 6 h as described subsequently. Recordings were performed on Grass Model 78 polygraphs calibrated to 50 J.l V/1O mm deflection and a paper speed of 10mm/s. Each animal receivedall three treatments, assigned in random order and separated by 1 wk each. Sleep recordings wereanalyzed by a single "blind" rater, who assigned a state of consciousness (waking, non-REM, or REM sleep) to each 30-s epoch according to standard criteria and definitions (10).
Respiration Respiration was assessed by the "barometric technique," which measures pressure differences between the test chamber containing the experimental animal and an empty reference chamber, both of which are ventilated with fresh room air (Model RM 80; Columbus Instrument Co., Columbus, OH). Small changes in pressure in the test chamber result from heating and cooling of the respired gas; their detection allows determination of respiratory rate in a totally noninvasive manner. Furthermore, there is a relationship between the magnitude of this signal and the tidal volume that has been well validated (11). In our device the pressure signal is AC coupled and then output to the polygraph for off-line measurement and calculation of tidal volume. A pressure pulse input into our chamber has a signal fall-time constant of 0.1 s. Preliminary testing of the chamber to determine its frequency response showed that at multiple signal sizes (different tidal volume equivalents) no change « 1%) could be detected in signal output as the frequency was varied from 50 to 150 Hz. Calibration was performed by pumping 7.55 ml/min of room air for 5 s into
the chamber using a Hagen Optima (Mansfield, MA) pump. A standard polygraph pen deflection representing approximately 0.5 ml was obtained; this was calculated from the height of the peak (1.55em) and the time from beginning of injection to peak deflection (0.065 s). Examples and descriptions of both sleep and respiratory recordings, as well as calibration procedures, are found in an earlier publication (12). It should be noted that the technique of calibration of volume used does not allow a determination of absolute volume that remains valid across individual animals but does allow comparison in one animal across treatments, especiallyif changes in volume are small.
Dosage oj Buspirone The doses were chosen on the basis of previous behavioral studies in the rat. We chose 10to 20 mg/kg given intraperitoneally based on the previous observation that 10 mg/kg increased wakefulness while decreasing REM and non-REM sleep (6). This is higher than the minimal dose previously found to suppress conflict behavior (1 mg/kg by way of mouth) (2) and the dose found to have anticonflict effects in the "thirsty rat confict test" (> 1 mg/kg by mouth) (13). It is less than the (Received in original form August 2, 1989 and in revised form November 10, 1989) 1 From the State University of New York, Stony Brook, New York; Rutgers University, Camden, New Jersey; and New York University, New York, New York. 2 Correspondence and requests for reprints should be addressed to Wallace B. Mendelson, M.D., Department of Psychiatry, State University of New York at Stony Brook, Health Sciences Center, TW, Stony Brook, NY 11794-8101.
1527
1528
MENDELSON, MARTIN, AND RAPOPORT
ED so for inhibition of conditioned avoidance response in rats (41 mg/kg by mouth) (2). There is some disagreement about the degree of behavioral sedation in the rat in this range. Weissmanand coworkers (13) found a decrease in spontaneous approaches to a drinking spout in an unpunished situation with > 2 mg/kg by mouth, but Riblet and coworkers (2) found that spontaneous motor activity is not reduced at doses lower than 100 mg/kg by mouth. These doses are significantly higher than those usually used clinically but seem relevant to the human situation since in the rat they have anticonflict effects with minimal or no sedation. Typical human dosages of buspirone used for anxiolytic therapy are 20 to 60 mg/day,
,---...
80
S
70
Q)
60
E
o
Vehicle
EZ2J 10 mg/ml 0020 mg/ml
1= 50 0Q) Q)
30
-8
20
I-
10
o
o
Fig. 1. Total sleep time following administration of vehicle and intraperitoneal buspirone, 10 and 20 mg/kg. ANOVA indicated significant effects for drug (df = 2, 18; F = 4.9; P < 0.02),time period (df 2, 18; F 29.6; P < 0.00001), and drug-time period interaction (df = 4, 36; F = 8.8; p < 0.0001). Significant post hoc contrasts showed that vehicle differed from 10 mg/kg (p < 0.01) and 20 mg/kg (p < 0.01) in the first 2 h.
=
40
U)
..L----'--~Ul......l"""'_
_ _.l..--l....J
WALLACE B. MENDELSON, JOSEPH V. MARTIN, and DAVID M. RAPOPORT
Introduction
Anxiolytic agents, such as the benzodiazepines, are among the most widely used drugs. In addition to concern about unwanted sedation in patients treated with many of these agents, it has been increasingly recognized that anxiolytics depress ventilation, particularly in patients with sleep-related respiratory impairment or chronic obstructive pulmonary disease (1). The azospirone buspirone is a nonbenzodiazepine that possesses anticonflict and antiaggressive effects in animals (2) and is anxiolytic in humans (3). Clinical reports have indicated that it is nonsedating (3, 4), and a polygraphic study found no significant effects of clinically used doses on the sleep of insomniacs (5), except to delay the onset and reduce the amount of rapid eye movement (REM) sleep. In rats, somewhat higher doses may actually decrease sleep (6). No effect of buspirone on the appearance of electroencephalographic (EEG) waveforms inside individual states of consciousness (waking, non-REM, and REM) has been reported. A preliminary clinical study has indicated that buspirone does not suppress ventilation (7). More recently,Garner and coworkers (8) have reported that buspirone stimulates respiration as measured by phrenic nerve responses in anesthetized, paralyzed cats whose carotid sinus nerves and vagi had been cut. To extend these observations to the intact and sleeping animal, we have examined the effects of two doses of buspirone on sleep and respiration in unanesthetized, freely moving rats. Method Sleep Recordings A group of 10male Sprague-Dawley rats (250 to 300g) wereanesthetized (70 mg/kg of ketamine and 6 mg/kg of xylazine, intramuscularly) and surgically implanted with 0-80 stainless steel screwsthat served as dural EEG electrodes; Teflons'-coated 0.010 inch stainless steel wires were embedded in the nuchal musculature for electromyographic recording (9). After a l-wk recovery period in chambers with lights on from 8:00 A.M. to 8:00 P.M.,
SUMMARY Drugs used in the treatment of anxiety are frequently sedating and tend to be respiratory depressants. Buspirone, a nonbenzodiazepine anxiolytic agent, has little reported sedative effect. It has been shown to be a respiratory stimulant in an anesthetized, glomectomized cat model. In this study, we examined the effects of two intraperitoneal single doses (10and 20 mg/kg) of buspirone on sleep and respiration in unanesthetized, intact, freely moving rats. Buspirone increased sleep latency (p < 0.0001)and decreased total sleep (p < 0.02) through reductions in both non-REM and REM sleep. Respiratory rate (p < 0.0003) and ventilation (p < 0.004) were significantly increased for 4 h after drug injection. The effects on respiration were independent of those on sleep; stimulation was evident in both waking and non-REM sleep. This study suggests that buspirone, in addition to being free of sedating and respiratory depressant side effects when prescribed for anxiety in humans, may be a respiratory stimulant whose effects persist in sleep. AM REV RESPIR DIS 1990; 141:1527-1530
animals were injected with vehicle (0.90/0 saline) or intraperitoneal buspirone, 10 or 20 mg/kg, at 9:00 A.M., after which sleep and respiration were recorded for 6 h as described subsequently. Recordings were performed on Grass Model 78 polygraphs calibrated to 50 J.l V/1O mm deflection and a paper speed of 10mm/s. Each animal receivedall three treatments, assigned in random order and separated by 1 wk each. Sleep recordings wereanalyzed by a single "blind" rater, who assigned a state of consciousness (waking, non-REM, or REM sleep) to each 30-s epoch according to standard criteria and definitions (10).
Respiration Respiration was assessed by the "barometric technique," which measures pressure differences between the test chamber containing the experimental animal and an empty reference chamber, both of which are ventilated with fresh room air (Model RM 80; Columbus Instrument Co., Columbus, OH). Small changes in pressure in the test chamber result from heating and cooling of the respired gas; their detection allows determination of respiratory rate in a totally noninvasive manner. Furthermore, there is a relationship between the magnitude of this signal and the tidal volume that has been well validated (11). In our device the pressure signal is AC coupled and then output to the polygraph for off-line measurement and calculation of tidal volume. A pressure pulse input into our chamber has a signal fall-time constant of 0.1 s. Preliminary testing of the chamber to determine its frequency response showed that at multiple signal sizes (different tidal volume equivalents) no change « 1%) could be detected in signal output as the frequency was varied from 50 to 150 Hz. Calibration was performed by pumping 7.55 ml/min of room air for 5 s into
the chamber using a Hagen Optima (Mansfield, MA) pump. A standard polygraph pen deflection representing approximately 0.5 ml was obtained; this was calculated from the height of the peak (1.55em) and the time from beginning of injection to peak deflection (0.065 s). Examples and descriptions of both sleep and respiratory recordings, as well as calibration procedures, are found in an earlier publication (12). It should be noted that the technique of calibration of volume used does not allow a determination of absolute volume that remains valid across individual animals but does allow comparison in one animal across treatments, especiallyif changes in volume are small.
Dosage oj Buspirone The doses were chosen on the basis of previous behavioral studies in the rat. We chose 10to 20 mg/kg given intraperitoneally based on the previous observation that 10 mg/kg increased wakefulness while decreasing REM and non-REM sleep (6). This is higher than the minimal dose previously found to suppress conflict behavior (1 mg/kg by way of mouth) (2) and the dose found to have anticonflict effects in the "thirsty rat confict test" (> 1 mg/kg by mouth) (13). It is less than the (Received in original form August 2, 1989 and in revised form November 10, 1989) 1 From the State University of New York, Stony Brook, New York; Rutgers University, Camden, New Jersey; and New York University, New York, New York. 2 Correspondence and requests for reprints should be addressed to Wallace B. Mendelson, M.D., Department of Psychiatry, State University of New York at Stony Brook, Health Sciences Center, TW, Stony Brook, NY 11794-8101.
1527
1528
MENDELSON, MARTIN, AND RAPOPORT
ED so for inhibition of conditioned avoidance response in rats (41 mg/kg by mouth) (2). There is some disagreement about the degree of behavioral sedation in the rat in this range. Weissmanand coworkers (13) found a decrease in spontaneous approaches to a drinking spout in an unpunished situation with > 2 mg/kg by mouth, but Riblet and coworkers (2) found that spontaneous motor activity is not reduced at doses lower than 100 mg/kg by mouth. These doses are significantly higher than those usually used clinically but seem relevant to the human situation since in the rat they have anticonflict effects with minimal or no sedation. Typical human dosages of buspirone used for anxiolytic therapy are 20 to 60 mg/day,
,---...
80
S
70
Q)
60
E
o
Vehicle
EZ2J 10 mg/ml 0020 mg/ml
1= 50 0Q) Q)
30
-8
20
I-
10
o
o
Fig. 1. Total sleep time following administration of vehicle and intraperitoneal buspirone, 10 and 20 mg/kg. ANOVA indicated significant effects for drug (df = 2, 18; F = 4.9; P < 0.02),time period (df 2, 18; F 29.6; P < 0.00001), and drug-time period interaction (df = 4, 36; F = 8.8; p < 0.0001). Significant post hoc contrasts showed that vehicle differed from 10 mg/kg (p < 0.01) and 20 mg/kg (p < 0.01) in the first 2 h.
=
40
U)
..L----'--~Ul......l"""'_
_ _.l..--l....J
