14,676-687 (1990)

FUNDAMENTALANDAPPLIEDTOXICOLOGY

Acute inhalation Toxicity of Soman and Sarin in Baboons A. ANZUETO,* R. A. DELEMOs,t H. HAMIL,~ D. JOHNSON,*

J. SEIDENFELD,* G. MOORE,+ AND S. G. JENKINSON*

*Depurtment qSMedicinr. University oj’Te.xas Health Science Center, tSouthwesc Foundation ,for Biomedical Research, and SSouthwest Research Institute. San .4ntonio. Texas

Received April 13. 1989: aceepled December 19, 1989 Acute Inhalation Toxicity of Soman and Sarin in Baboons. ANZUETO, A., DELEMOS, R. A., J.. MOORE, G., HAMIL, H., JOHNSON, D.. AND JENKINSON, S. G. (1990). Fundam. Appl. Tosicol. 14,676-687. Adult baboons (Papio sp.; 8-12 kg) were anesthetized with sodium pentobarbital (20 mg/kg iv). The animals were instrumented for measurement of mean blood pressure (MBP), pulmonary artery pressure (PAP), ECG, arterial and mixed venous blood gases, lung volumes, lung pressures, and efferent phrenic nerve activity. Bronchoalveolar lavage (BAL) was performed. Studies were done prior to exposure, at intervals during the first 4 hr postexposure, and at 4 and 28 days after exposure. Control animals received a sham exposure to 2propanol (N = 5). Soman (pinacolyl methylphosphonofluoridate) at 13.14 @g/kg (2 X LD50) was vaporized into the upper airway in a second group of animals (N = 5), and sarin (isopropyl methylphosphonofluoride) 30 pg/kg (2 X LD50) was vaporized into a third group of animals (N = 4). Controls showed no change in any parameter either immediately after diluent exposure or during the monitoring period. Soman and sarin produced a decline in MBP and bradyarrhythmias that were reversed with atropine. Apnea occurred in all soman- and sarin-exposed animals within 5 min postexposure, and was associated with absence of phrenic nerve signal. Ventilation was mechanically supported until the animal could maintain normal arterial blood gasesduring spontaneous breathing. BAL studies revealed an increase in total white cell population and neutrophils at 4 hr in all three groups. There were signs of impaired hemodynamics and persistent lung injury for 4 days that resolved by 28 days after exposure. In conclusion, inhalation of soman and sarin in the baboon is associated with cardiac arrhythmias, development of apnea, and a significant decrease in MBP. Inhalation exposure also resulted in a persistent influx of neutrophils and hypoxemia. 8 1990 Society ofToxicology. SEIDENFELD,

Intoxication with organophosphorus cholinesterase inhibitors used as insecticides constitutes an increasing problem in industrial and agricultural toxicology. Similar compounds, e.g., soman and sarin, are considered as potential warfare agents due to their toxicity. These compounds produce a wide variety of toxic manifestations (Goodman et al., 1985). Although studies using many different species have been published concerning the pharmacological effects of such compounds following various routes of injection (Anzueto et al., 1986; Doebler et al., 1984; Gupta et al., 1987; Lipp and Dola, 1980; Rickett et al., 1982; Wolthuis and Meeter, 1968), fewer 0272-0590/90 $3.00 Copyright 8 1990 by the Society ofToxicology. All rights of reproduction m any form reserved.

676

experiments have been published on inhalation exposure (Aas et al., 1985; Crook et al., 1969; Fonnum et al., 1984; Fredriksson et al., 1960). In addition, there are no comparative data between inhalation and parenteral exposure in the literature using nonhuman primates. We have developed a nonhuman primate model to study the effects of organophosphorus compounds, and used this model to define the pathophysiology of soman and sarin following intravenous (iv) injection (Anzueto et al., 1986; Berdine et al., 1985; Johanson et al., 1985). These studies demonstrated that the pathophysiologic response in the baboon

INHALATION

OF

closely resembles that in humans and other animal models. In this report, the first objective was to determine the effects of inhaled soman and satin on hemodynamics and lung function in baboons, including study of cellular constituents of the lower respiratory tract by bronchoalveolar lavage. Our second objective was to compare these effects with those of previous studies using iv exposure in the same animal model (Anzueto et al., 1986). An accurate comparison of the two routes of exposure in the same animal model, with the same instrumentation and methodology, would provide better understanding of toxic and other effects of these agents. In summary, we found that inhalation of soman and sarin produced a decline in blood pressure and bradyarrhythmias that were reversed with atropine. All the animals exposed developed apnea that was preceded by severe arterial hypoxemia. Inhalation exposure also resulted in a persistent influx of neutrophils and hypoxemia. METHODS Animal Model Animal National

care was conducted under the guidelines ofthe Research Council’s Guidefir the Care and U.se qflaboratory Animals. All protocols were reviewed and approved by the Animal Research Committee of the Southwest Foundation for Biomedical Research. This experimental baboon model has been described in detail previously (Anzueto et al.. 1986). In brief, adult baboons (Papio cynocephalus anubis). 8- 12 kg, were anesthetized with iv sodium pentobarbital (initial dose 20 mg/kg). This initial dose was supplemented as required to maintain adequate anesthesia. The level of anesthesia was regulated to keep the cornea1 reflex abolished. This index of anesthetic depth was checked every IO- I5 min during the study. Animals were instrumented with pulmonary and systemic arterial lines, urinary catheters, subcutaneous ECG electrodes, and pleural and abdominal latex balloon-tipped catheters. A root of the phrenic nerve was bilaterally isolated in the neck, and a set of bipolar platinum electrodes was positioned. The upper airway of all the animals was fitted with a mouth mold through which a large-bore tubing was passed. The end of the tubing was adjusted so that it rested on the dorsum of the tongue well above the epi-

SOMAN

AND

SARIN

677

glottis. The tube was connected to a clear, rigid, snouttapered shell and sealed by a closed-cell urethane foam. The purpose of this device was to provide a patent upper airway with minimal dead volume and an attachment for the vaporizer delivery system, and to maintain an airtight seal of the upper airway so satisfactory measurements of airflow at the mouth were possible. Bronchoalveolar lavage (BAL) was performed by a technique previously described by our group in baboons (Higuchi et al., 1982). The recovered fluid was collected in two 50-m] polypropylene centrifuge tubes, strained through cotton gauze, placed on ice, and then centrifuged at 25OOg for 10 min at 4°C to pellet the cells and to produce a cell-free supernatant. A 20-~1 sample of cell suspension obtained after centrifugation was treated with hemolysin, and a cell count performed with a hemacytometer on four large squares. Viability was checked by trypan blue exclusion with 100 cells counted. A 200-J aliquot of the cell suspension was removed for centrifugation and stained with Wright stain. A differential cell count was performed on 200 cells. The technique of Pesanti (1979) was modified and used to study phagocytosis by alveolar cells. Leukocyte preparations were adjusted with minimal essential medium (MEM + 10% calf serum), such that 0.25-m] aliquots containing 5 X IO5 leukocytes could be plated in duplicate on 16-mm wells (24well plates, Co-star. Cambridge, MA). brought to 0.5 ml with MEM, and incubated I hr at 30°C in a humidified 5% CO* environment. To each sample, 2 X 10’ washed 3H-labeled Staphylococcus aweus in 0.25 ml of MEM were added at a final bacterium-to-cell ratio of 40 colonyforming units (CFU)/leukocyte). After further addition of 0.25 ml of MEM, the mixtures were incubated 1 hr as before. Supernatants were then aspirated and the cell monolayers were incubated with 50 mg of lysostaphin (200-300 units/mg; Sigma) in 2 ml of MEM for 15 min. After being rinsed twice in PBS, leukocytes were ruptured by overnight incubation at room temperature with 0.5 ml of 0.1 N NaOH/well. Contents of each well were added to IO ml scintillation cocktail containing 0.5 ml of 0.1 N HCI and counted in a Beckman LS-7500 liquid scintillation counter (Beckman Instruments, Fullerton, CA). Counts were converted to disintegrations per minute (dpm) based on a standard curve established for the machine used, and counts were corrected for the number of organisms added to each well, the number of leukocytes per well, and the number of dpm per CFU. Routine blood studies included complete blood count (CBC). platelet count, and biochemical and electrolyte profiles (glucose, BUN, creatinine, total protein, albumin/globulin ratio, SGOT, SGPT. alkaline phosphatase. sodium, potassium, chloride, total CO& and acetylcholinesterase (AchE) activity was determined in both red cells and serum. The “‘C radiochemical procedure used by Sterri et al. ( 1979) was used to measure acetylcholinesterase activity in plasma and red cell samples. For each animal. whole blood was collected prior to exposure and at 4 hr. 4 days, and 28 days after exposure. The blood was

678

ANZUETO

ET AL.

centrifuged to separate the plasma and red cell fractions. The fractions were stored at -70°C until analysis on the third day. Analytical data expressed as mmol/ml/min were normalized to the percentage change from preexposure data.

agent delivery system consists of a small sintered stainless-steel heating element into which the required quantity of dilute agent is absorbed and converted to a vapor by application of electric current to the heating element.

Physiological Measurements

Experimental Protocol

Arterial and pulmonary artery (PA) pressures. ECG, and heart rate were recorded continuously throughout the experiment. Pulmonary artery wedge (P,,) pressures were obtained at intervals by inflating the PA catheter until a tracing representing the pulmonary artery occlusion pressure was obtained. Lung water was measured using the double-indicator dilution technique, described by Gray et al. (I 984). An injection of IO ml of iced indocyanine green (Cardiogreen) was made through the proximal port of the PA catheter. Blood was withdrawn at a constant rate using a withdrawal pump from the thermistor catheter in the distal aorta. This blood was aseptically passed through a densitometer cuvette and returned to the animal at the end of the withdrawal period. Lung water was calculated by computer analysis of the transit time of heat, as sensed by the thermistor, and Cardiogreen as measured by the densitometer. Cardiac output (CO) was measured by thermal dilution using a 3-ml DSW injectate. These measurements were made in triplicate and averaged. Lung volumes (inspiratory capacity, functional residual capacity, total lung capacity), lung diffusing capacity (DL,,), diffusing capacity per unit lung volume (DLVA). and lung resistance were measured using the techniques described by Felton and Johanson (I 980) Sackner ef al. (1975) and Pengelly ( 1977). The mechanical performance of the diaphragm was monitored by measuring the transdiaphragmatic pressure (Pdr) during spontaneous breathing, as well as after indirect muscle stimulation (through the phrenic nerve). The technique for standardizing diaphragm contraction was to measure Pdi during bilateral phrenic nerve stimulation using the techniques described by Aubier et al. ( I98 I. 1985). The phrenic nerves were stimulated using a Grass stimulator (Grass Instruments, Quincy, MA). These data were analyzed by plotting the Pd, generated against the frequency of phrenic stimulating signal (force/frequency relationship).

On the exposure day. baseline measurements were performed, which included hemodynamics (MBP, CO, PA, and Paw), lung volumes, lung diffusing capacities, Pdi, arterial blood gases, mixed venous blood gases, and indirect diaphragm stimulation; then the animal’s head was introduced into a surety hood and connected to the inhalation delivery device. A group of animals (N = 5, group A) received vaporized solvent 2-propanol and served as controls. Soman 13.14 &kg (2 X LD50) was vaporized into a second group of animals (N = 5, group B), and satin 30 pg/kg (2 X LD50) was vaporized into a third group ofanimals (N = 4, group C). These doses of soman and sarin were chosen based on our previous experience (Anzueto et al., 1986; Berdine et al., 1985, Johanson et al., 1985) that. in the baboon model, 2 X LD50 ofsoman or satin produced severe hemodynamic changes, which were reversed with atropine. and apnea, which required mechanical ventilation for 2 to 4 hr. The agent was delivered in two similar doses approximately I min apart. The expired breath of the animal was monitored with an impinger train for 3 min from the initial exposure. Preliminary pilot studies indicated that when the total dose was delivered in a single breath, it resulted in a cough and subsequent loss of the agent in the expired air and the collection system. The exposure device and mouthpiece were removed from the animals approximately 3 min after the initial exposure to allow intubation and mechanical ventilation. Continuous clinical evaluations were performed for secretions, muscle fasciculation, and seizure activity. At the onset ofapnea, the animals were endotracheally intubated and ventilated by a volume ventilator (MA- I, Puritan-Bennett Corp.. Kansas City. KS). Atropine 1.5 mg was given im when the animal exhibited signs of intoxication (i.e.. cardiac arrhythmias, hemodynamic changes, muscle activity, apnea). A second dose was administered as clinically indicated. The animals were continuously monitored and ventilated until they were able to sustain spontaneous breathing and all other systems were stable. This period averaged 4 hr. At this point, instrumentation was removed and the animals were returned to cages. All hemodynamic, pulmonary function, bronchoalveolar lavage, and blood chemistry studies were repeated 4 and 28 days after the agent exposure. The monitoring intervals of 4 and 28 days postexposure were chosen to evaluate the extent of the acute effects and to identify long-lasting abnormalities, respectively. The instrumentation described previously was inserted and removed each monitoring period.

inhalation Delivery Sy.ytem The conceptual design of the inhalation device and the agent delivery system inside a baboon’s mouth is shown in Fig. I. This inhalation delivery system has been reported in detail previously (Johnson et al.. 1988; Hamil e( al., 1986). In brief, this system consists of a breathing tube with inlet and exit valves that passes through a custom-fitted mouthpiece inside the animal’s mouth. The

INHALATION

OF

SOMAN

AND

679

SARIN

POWER CONTROLLER

AGENT DELIVERY

TUBS

rrz\A\

AGENT DELIVERV

q

FIG. 1. Drawing mask in an animal.

of the agent delivery

SYRINGE

POWER, LSAO

system

and the inhalation

Orgunophosphate Agent Stock solutions of dilute soman and satin were obtained from the U.S. Army Research Medical Institute of Chemical Defense (USAMRICD), Edgewood. Maryland.

device

inside

the ventilatory

support

tween groups by Dunnet’s repeated-measures comparison test. Statistical calculations were performed as described by Zar (I 974) and Winer (197 1); the criterion for statistical significance was set at p < 0.05.

RESULTS Statistical Anaivsis and Data Interpretation Mean. standard deviation, variance, and standard error of the mean were calculated for each variable on grouped data for all animals within a group. The BAL data were analyzed using regression techniques to describe the rate and extent of changes and analysis of variance (ANOVA) techniques to detect differences between treatment groups. Following ANOVA, significant differences between groups were assessed using a Student/ Neumann-Keuls (SNK) test. Analyses of other variables having more than two determinations were performed with a repeated-measures experimental design. Postexposure means were compared to baseline means and be-

The control animals (group A) showed no significant change over time in any of the physiological parameters monitored. None of the animals had any cardiac arrhythmias or other hemodynamic abnormality. Although changes in the respiratory pattern related to the level of anesthesia were noted, all the animals had spontaneous ventilation throughout the experiment. After the sham exposure to 2-propanol, there were no changes in the breathing pattern or cough response.

680

ANZUETO

ET AL.

TABLE I AMOUNTSOFSOMANANDSARINLOADEDANDDELIVEREDTOEXPOSEDANIMALS

Animal

Weight (kg)

Dose loaded bd

Impinger catch w

Dose delivered rg

9%

159.79 148.80 145.50 121.40 129.50

99.8 95.4 99.7 93.4 90.0 141 f 15

283.6 286.2 250.7 363.1

93.3 92.9 95.0 99.2 296 2 48

(a) Soman-exposed group 1 2 3 4 5

11.7 11.3 10.9 9.7 10.5

160 156 146 130 144

0.21 1.20 0.50 8.60 14.50 Mean f SD

(b) Satin-exposed group 1 2 3 4

9.9 10.1 8.8 12.2

304 308 264 366

20.4 21.8 13.3 2.9 Mean k SD

Table I shows doses of soman and sarin loaded for each animal and the dosage amount delivered. In the soman-exposed animals, the mean dose delivered was 14 1 + 15 mg, 96% of the loaded dose; and the sarin exposure was 296 * 48,95% of the loaded dose. General effects of soman and satin intoxication included mild muscle fasciculations but none progressed to generalized seizurelike activity. This finding was expected because the animals had been anesthetized with pentobarbital. We have previously noted the suppressive effect of pentobarbital on muscle fasciculation and seizure activity after soman exposure (Anzueto et al., 1986). All exposed animals also had increased oral, bronchial, and gastric secretions. The secretions were noted at the onset of apnea, then decreased after atropine administration, and in some animals were prominent again 4 hr after agent exposure. Table 2 shows the AchE activity as the percentage change from preexposure. The control animals showed some variation in plasma AchE activity, mainly on days 4 and 28, but not in RBC AchE activity. Both soman- and satin-exposed animals

showed a significant decrease in AchE 4 hr after exposure; AchE remained decreased in RBC at 4 days. All soman- and sarin-exposed animals developed bradyarrhythmias, including first-, second-, and third-degree heart blocks. These TABLE 2 PERCENTAGECHANGEINACETYLCHOLINESTERASE ACTIVITYAFTEREXPOSURE' Time after exposure Group Control (n = 4) Plasma AchE RBCb AchE Soman (n = 5) Plasma AchE RBC AchE Satin (n = 4) Plasma AchE RBC AchE

4 days

28 days

-27 * 10 -7f 5

-25 31 14 -7f 6

-86 f 2* -93 F 2*

-44k -86i

8 4*

-34k -54k

-81 i 3* -91* I*

-33k -85 f

6 2*

-3* 2 -35 f 11

4 hr f4f3 +5?2

a Mean k SEM. b RBC, red blood cells. * p < 0.0 1 compared with control group by ANOVA.

9 12

INHALATION

30 , Baseline

OF SOMAN

l I

Apnea

I

14

681

AND SARIN

I

I

I

30

60

120

/ 240

?y 4

Day 28

TIME FIG. 2. Comparison of the mean blood pressure (MBP) response in control, soman, and sarin groups. Time is expressed first in minutes, then in days, after exposure. Open circles are the control animals (N = 5); closed circles are the animals exposed to soman, 13.14 fig/kg (2 X LD50, N = 5); and open triangles are the animals exposed to satin, 30 pg/kg (2 X LD50, N = 4). Atropine, 3 mg im, was injected into both soman and satin animals. Bars indicate GE. *Data points significantly different from preexposure (p < 0.05) by two-tailed Dunnet’s multiple-comparison test. **Data points significantly different from control (p C 0.05) by two-tailed Dunnet’s multiple comparison test.

arrhythmias occurred early in the intoxication, occasionally as the initial manifestation, and responded promptly to atropine. The mean blood pressure changes are shown in Fig. 2. Blood pressure fell in all soman- and sarin-exposed animals at the same magnitude, showing its lowest value at the time of apnea. Blood pressure improved slowly after administration of atropine and remained below 90 mm Hg during the first 1 hr following agent exposure. Mean pulmonary artery pressure changes in soman-exposed animals showed an increase from a mean value of 16 k 2 mm Hg at baseline to 19 ~fr3 mm Hg at apnea. This change was not statistically significant. The sarin-exposed animals showed an increase from a mean value of 16 f 0.9 mm Hg at baseline to 23 2 2.2 mm Hg at 8 min postexposure at apnea. Pulmonary artery pressure returned to baseline levels within 60 min of exposure. No change in Paw occurred throughout the experiment. Cardiac output fell in all soman- and sarin-exposed animals after agent inhalation, but im-

proved after atropine administration. Other hemodynamic parameters showed the same changes after soman and sarin administration and responded to atropine injection. In all animals, the extravascular lung water was measured using the double-indicator dilution technique. There were no significant changes in lung water values. No major complication of the procedure occurred and none of the animals developed hypothermia or hemorrhagic problems. The initial respiratory response to inhaled soman and sarin was usually tachypnea with a progressive decrease in tidal volume. Paradoxical inward movement of the abdomen during respiration and alternation between abdominal and thoracic breathing were prominent clinical features in all animals. The animals progressed rapidly to apnea manifested by the absence of regular respiratory efforts; however, during this phase, the animals made irregular respiratory efforts which resulted in large tidal volumes. Apnea was present, on average, 5 f. 1.5 min after

682

ANZUETO

ET AL.

CE ; 80E 0” B

60-

40Baseline

Apnea

14

30

60

120

240

Day

Day

TIME FIG. 3. Comparison ofthe arterial oxygen pressure (P,O,) response in control, soman, and satin groups. Time is expressed frst in minutes, then in days, after exposure. Open circles are the control animals (N = 5); closed circles are the animals exposed to soman, 13.14 &kg (2 X LD50, N = 5); and open triangles are the animals exposed to satin. 30 fig/kg (2 X LD50, N = 4). Atropine, 3 mg im, was injected in both soman and satin animals. Bars indicate *SE. *Data points significantly different from preexposure (p G 0.05) by two-tailed Dunnet’s multiple comparison test. **Data points significantly different from control by two-tailed Dunnet’s multiple comparison test.

agent exposure. All of the animals in groups B and C were intubated and placed on a volume ventilator. Arterial blood gases did not detetiorate until the onset of apnea. Figure 3 shows P,O, response at the time animals required ventilatory assistance: P,Oz, 42 5 7 mm Hg, and P,C02, 59 f 5 mm Hg, in the soman-exposed animals; P,Oz, 47 t 11 mm Hg, and P&O,, 52 I~I 6 mm Hg, in the sarin-exposed animals. P,Oz remained above 75 mm Hg over the entire period of ventilatory support in both groups of animals. By 4 hr postexposure, all the animals were able to maintain adequate gas exchange during spontaneous breathing, with P,Oz within baseline values. The diaphragms of both soman- and sarinexposed animals were stimulated through the phrenic nerve to assessthe state of the neuromuscular junction and the ability of the muscle to generate force (transdiaphmgmatic pressure, Pdi). Soman-exposed animals showed no change in P,,, at apnea, but there was a progressive decline in the pressure response over the subsequent 4 hr; by Day 4, Pdi was within

70% of baseline. The satin-exposed animals had a P,+ within 70-80% of baseline in the postexposure period; this returned to baseline values at Day 4. None of these changes were statistically significant. The effect of inhaled soman on lung function was primarily manifested by an increase in lung resistance. Resistance increased in the soman-exposed animals (group B) from a mean value of 2.08 + 0.5 cm HzO/liter/min at baseline to 3.67 + 0.53 (176% increase) at 30 min, 6.38 + 2.08 cm HzO/liter/min (307% increase) at 3 hr, and 5.97 f 1.6 cm H,O/liter/min (287% increase) at 4 hr after agent exposure. Other parameters such as lung volume and lung diffusing capacity showed no significant change. Bronchoalveolar lavage data showed no significant differences in total cell count and viability analyzed by two-way analysis of variance with group and time as variables. All groups had significant decreases in the percentage of alveolar macrophages and increases in the percentage of neutrophils 4 hr after inhalation. The soman-exposed animals

INHALATION

OF SOMAN

683

AND SARIN

TABLE 3 CELLULARCHARACTERISTICSOFBRONCHOALVEOLARLAVAGEFROMBABOONSEXPOSED BYINHALATIONTOSOMANANDSARIN

Group

Baseline

Hour 4

Day 4

Day 28

Pulmonary alveolar macrophages (W) Control Soman Sarin

60-+21 81 f 10 82i 8

36 f 36* 46 f 20* 6Ok 17*

58+ 13 32 k 7** 59 * 12

90* 86k 85k

26k 17 62 f lo** 34* 17

19f 17 llf 6 4-r- 2

3 6 7

Neutrophils (%) Control Soman Sarin

4* 3 5k 3 Sk 6

51 -t40* 46 + 25” 31 +- 16*

* p < 0.05 different from other time points. ** p Q 0.05 different from other groups at Day 4 and other time points except Hour 4 in the other three groups by one- and two-way analyses of variance.

continued to show depression of macrophages and elevation of neutrophils 4 days after exposure, whereas the others returned to normal values. There were no significant differences in phagocytosis values after ingestion of Staphylococcus aureus either by group or by time, and there was no significant group-time interaction. The means + SD for individual groups and individual time points are shown in Table 3. DISCUSSION We have characterized the physiologic consequences of soman and sarin inhalation exposure in the nonhuman primates. There were acute hemodynamic effects that were partially corrected after atropine administration. The exposed animals showed severe irritation of the upper airway, followed by hypoxemia and apnea. The animals required mechanical ventilation until recovery, and by 4 days postexposure, both hemodynamic and ventilatory parameters had returned to baseline. The inhalation system used provided repeatable and quantitative agent administration to individual test animals; but by using this inhalation system, there are various pos-

sible sources of errors, mainly in estimating the total amount of agent delivered. Accurate measurement of the absorbed dose depended mainly on accurate measurement of the quantity of agent exhaled. Two possible sources of error were the difficulty in determining the bubbling chambers’ capacity to absorb all exhaled agent that passed through them and the failure to estimate the amount of agent adhering to the chamber walls or tubing prior to reaching the bubbling chamber. Based on the animals’ physiologic response and AchE inhibition, we believed that the animals received most of the delivered dose and the amounts of exhaled agent were not significant. We were also concerned that this inhalation system bypassed the nose. Nasal passages are known to protect the lower airway by warming, humidifying, and clearing inspired air. Exposure of the nose to toxic particulate matter (Torjusseer, 1983) and irritant gases (Buckley et al., 1984), results in systemic absorption of the irritant and lesions in the nasal mucosa. Once an irritant has bypassed the nose, nerve endings concentrated in the carina and large airways can stimulate cough and help to remove irritants that reach the lower respiratory airway (Widdicombe, 1954).

684

ANZUETO

30

,

I

I

ET AL.

I

I

0

TIME

I

I

, 240

fmin)

FIG. 4. Comparison of the mean blood pressure (MBP) response in iv and inhalation soman-exposed groups. Closed circles are animals exposed iv to soman. 13.14 pg/kg (2 X LD50, N = 6); and open triangles are animals exposed by inhalation to soman, 13.14 fig/kg (2 X LD50. N = 5). Atropine, 3 mg im. was injected into both soman groups. Bars indicate *SE. *Data points significantly different from preexposure (p < 0.05) by two-tailed Dunnet’s multiple-comparison test.

Most primates and human subjects breath through the nose at rest, but Niinimaa et al. ( 1980) demonstrated that adult subjects switch to oronasal breathing at higher minute ventilation. Furthermore, Lippman et al. ( I97 I), using radioactive particles, found that deposition in the upper airway of mouthbreathing subjects increased with increased respiratory rate due to impaction of inhaled particles in the back of the mouth. In our experimental animals, the initial response to soman and sarin was cough, followed by a marked increase in respiratory rate and tidal volume. Based on these observations we concluded that bypassing the nose with the inhalation system did not produce any alterations in the animals’ response. Furthermore, this response was similar to the clinical observations described in other animal models (Aas et al., 1985; Crook et al., 1969; Fonnum et al., 1984). The inhalation system used offers safety and quantitative delivery of toxic compounds that outweigh the possible limitations.

Data from Aas et al. ( 1985) on rodents exposed to inhaled soman showed that the sequence of gross signs of intoxication was the same as if soman were administered by iv injection. Salivation was an early sign. Generalized muscle fasciculations were rapidly followed by convulsions. These findings were similar to those reported in dogs by Fredriksson et al. (1960). We could not directly compare inhalation data from these primate studies with existing inhalation information from rodents. Our experimental animals received a lethal dose of the agent, followed by intensive resuscitative support. The impact of resuscitation and supportive treatment clearly resulted in a clinical pattern different from that described in the more limited rodent studies. Comparison of these experiments on baboons with previous experiments on a similar model exposed intravenously (Anzueto et al., 1986) shows that the signs of intoxication are the same and are independent of the route of administration. Both groups of animals showed early hemodynamic changes charac-

INHALATION

OF SOMAN

AND SARIN

685

TIME (min) FIG. 5. Comparison of the arterial oxygen pressure (P,O,) in IV and inhalation soman exposed groups. Closed circles are animals exposed IV to soman, 13.14 &kg (2 X LD50, N = 6); and open triangles are animals exposed by inhalation to soman, 13.14 pg/kg (2 X LD50, N = 5). Arrows indicate time of start to mechanical ventilation in both soman groups. Bars indicate *SE. *Data points significantly different from preexposure (p G 0.05) by two-tailed Dunnet’s multiple-comparison test. **Data points significantly different from iv exposure by two-tailed Dunnet’s multiple-comparison test.

terized by cardiac arrhythmias, increased pulmonary artery pressure, decreased cardiac output and blood pressure, followed by muscle tremor, severe hypersalivation, and apnea. Figure 4 compares the blood pressure effects over a 4-hr period after inhalation or iv soman exposure. There was a significant decrease in MBP in both groups at apnea compared to baseline, but there were no significant differences between groups. Comparison of other hemodynamic effects in animals exposed intravenously and by inhalation to soman and satin showed the same relationship. Figure 5 compares the arterial oxygen concentration (P,O,) in soman-exposed animals after both inhalation and iv exposure. The inhalation-exposed animals showed more severe hypoxemia that was significantly different at apnea compared with the iv group. One possible explanation for this important difference is the presence of more severe bronchoconstriction in the inhalation-exposed animals. The bronchoconstriction may have been elicited by stimulation of irritant receptors by the agent. Clinical observations

in these inhalation-exposed animals showed hyperemia of the upper airway acutely after exposure, and in previous pilot studies the delivery of high doses of soman and sarin by inhalation produced a severe cough reaction. Stimulation of irritant receptors is known to cause bronchoconstriction by increasing stimulation of vagal efferent activity (Barnes, 1986; Laitinen, 1985; Sampson and Vidruk, 1975) or indirect release of inflammatory mediators (Barnes, 1986; Sampson and Vidruk, 1975). The baboons exposed to sarin by inhalation showed the same clinical signs of intoxication reported after accidental human exposures to sarin (Grob, 1956; Grob and Harvey, 1958). In human subjects, sarin produced local tissue effects such as mucosal irritation and cough, increased secretions, tightness in the chest with prolonged wheezing expiration, dyspnea, and bradycardia. Bronchoalveolar lavage provides a relatively noninvasive approach to the sampling of the alveolar milieu. To our knowledge, there are no published data on BAL changes after organophosphorus intoxication. In our experiments, the two experimental groups of

686

ANZUETO

animals showed the same changes in percentages of white cells, alveolar macrophages, and polymorphonuclear leukocytes acutely after the agent exposure. But the soman-exposed animals had significant persistent changes 4 days after exposure. Polymorphonuclear leukocytes were related to both instrumentation and manipulation of the animal and to the inhalation procedure. Persistence of polymorohonuclear leukocvtes cells in the soman-exposed animals,* however, suggests some agent-related lung injury with this agent and dose. Our data seem to confirm that adequate supportive treatment in the acute phase after agent exposure is essential in obtaining longterm survival. It is important to note that we studied these animals at fixed time intervals, and we have no physiological data to evaluate between the experimental intervals. We noted that 2 or 3 davs after exnosure a few animals exhibited increasing salivation, but no other clinical signs of intoxication were present. Our findings did confirm that spontaneous recovery of central respiratory function after intoxication with soman and sarin may not be related to the return of AchE activity (Adams et al., 1976). In summary, we have developed a reliable inhalation delivery system, and at the same time, we have characterized the physiologic consequences of soman and sarin inhalation exposure in the nonhuman primate. There are some differences in the time course of toxic manifestations between inhalation and parenteral exposure, in particular, irritation of the upper airway and severe hypoxemia after inhalation. The hemodynamic effects were similar in severity and duration. Once the animals have recovered from the acute phase, there are signs of persistent lung injury and hemodynamic effects for 4 hr that resolve by 4 and 28 days after exposure. We feel that this animal model can be utilized to develop an optimal strategy for protection and/or treatment of acute organophosphorus exposure. d

1

ACKNOWLEDGMENTS We thank Wesley Cox and Joseph Biewer for their excellent technical assistance, and Toya Harris for assis-

ET AL.

tance in the preparation of this manuscript. This work was supported by the U.S. Army Medical Research Institute of Chemical Defense (USAMRICD), under Con-

tract DAMD,7-85-C-5 159,

REFERENCES AAS, P., STERRI, S. H.. HJERMSTAD, H. P., AND FONNUM, F. (1985). A method for generating toxic vapors of soman: Toxicology of soman by inhalation in rats. Toxicol.

Appl. Pharmacol.

80,437~445.

ADAMS, G. K., III. YAMAMURA, H. I., AND O’LEARY, J. F. (1976). Recovery of central respiratory function following anticholinesterase intoxication. Eur. J. Pharmacol. 38, 10 1- I 12. ANZUETO, A., BERDINE, G. G., MOORE. G. T., GLEISER, C., JOHNSON, D.. WHITE, C. D.. AND JOHANSON, W. G., JR. (1986). Pathophysiology ofsoman intoxication in primates. Toxicol. Appl. Pharmacol. 86,56-68. AUBIER, M.. FARKAS, G., DE TROYER, A., MOZES, R., AND ROUSSOS,C. (198 1). Detection of diaphragmatic fatigue in man by phrenic stimulation. J. Appl. Physiol. 50,538-544.

AUBIER, M., MURCIANO, D.. LECOCGUIC, Y., VIIRES. N., AND PARIENTE, R. (1985). Bilateral phrenic stimulation: A simple technique to assessdiaphragmatic fatigue in humans. J. Appl. Physiol. S&58-64. BARNES. P. J. (1986). Neural control of human airwavs in health and disease. Amer. Rev. Resp. Dis. 13i, 1289-1314. BERDINE, G. G., ANZUETO, A., MOORE, G. T., WHITE, C. D., AND JOHANSON, W. G.. JR. (I 985). Etiology of respiratory failure in organophosphate intoxication. Amer. Rev. Resp. Dis. 131(Suppl.), A297 (Abstract). BUCKLEY, L. A., JIANG, X. Z., JAMES, R. A., MORGAN, K. T., AND BARROW, C. S. (1984). Respiratory tract lesions induced by sensory irritants at the RD50 concentration. To.xicol. Appl. Pharmacol. 74,4 17-429. CROOK, J. W.. MUSSELMAN, N. P., HESS, T. L., AND OBERST, F. W. ( 1969). Acute inhalation toxicity of difluoro vapor in mice, rats, dogs, and monkeys. Toxicol. Appl. Pharmacol. 15, 13 I-135. DOEBLER. J. A., WALL, T. J.. MOORE, R. A., MARTIN, L. J.. SHIH, T.-M., AND ANTHONY, A. (1984). Cytophotometric analyses of brain neuronal RNA in soman intoxicated rabbits. Toxicology 32, I53- 163. FELTON, C. R., AND JOHANSON, W. G., JR. (1980). Lung tissue volume during development of edema in isolated canine lungs. J. Appl. Physiol. 48, 1038-1044. FONNUM, F., PAAL, A., SIGRUN, S., AND HELLE, K. ( 1984). Modulation of the cholinergic activity of bronchial muscle during inhalation of soman. Fundam. Appl.

Toxicol.

4, S52-S57.

FREDRIKSSON,T.. HANSSON, C. H., AND HOLMSTEDT, B. (1960). Effects of sarin in the anesthetized and unanesthetized doa followina inhalation. oercutaneous 1 absorption and intravenous infusion. Arch. Int. Pharmacodyn.

Thu.

126,288-302.

INHALATION

OF SOMAN

GOODMAN, A., GOODMAN, L. S., RALL, T. W., AND MURAD, T. (1985). The Pharmacological Basis of Therapeutics, 7th ed. MacMillan, New York. GRAY, B. A., BECKETT, R. C., ALLISON, R. C., MCCAFFREE, R., SMITH, R. M., SIVAK, E. D., AND CARLILE, P. V., JR. (1984). Effect of edema and hemodynamic changes on extravascular thermal volume of the lung. J. Appl. Physiol. 56,878-890. GROB, D. (1956). The manifestations and treatment of poisoning due to nerve gas and other organic phosphate anticholinesterase compounds. Arch. Intern Med. 98,22 l-239. GROB, D.. AND HARVEY, J. C. (1958). Effects in man of the anticholinesterase compound sarin (isopropyl methyl phosphonofluoridate). J. Clin. Invest. 37,350368.

GUPTA, R. C., PATTERSON, G. T., AND DETTBARN, W-D. (1987). Biochemical and histochemical alterations following acute soman intoxication in the rat. Toxicol. Appl. Pharmacol. 87,393-402. HAMIL, H., ANZUETO, A., JOHNSON, D., DELEMOS, R.. SEIDENFELD, J. J., AND MOORE, G. (1986). A system for generating toxic vapors of organophosphate nerve agents. In Proceedings. Sixth Annual Chemical Defense Bioscience Review. HIGUCHI, J. H.. COALSON, J. J., AND JOHANSON, W. G.. JR. (1982). Bacteriologic diagnosis of nosocomial pneumonia in primates. Usefulness of the protected specimen brush. Amer. Rev. Respir. Dis. 125,53-57. JOHANSON, W. G., JR., ANZUETO, A. A., BERDINE, G. G., MOORE, G. T., AND WHITE, C. D. (1985). Etiology of respiratory failure in organophosphate intoxication. In Proceedings, 5th Annual Chemical Defense Bioscience Review. JOHNSON, D. E., ANZUETO, A., HAMIL, H., BREWER, J. H., MOORE, G. T., SEIDENFELD, J. J., AND DELEMOS, R. A. ( 1988). Studies of the Eficts of Organophosphorous Agent Exposure on the Lung, technical report. U.S. Army Medical Research and Development Command, Fort Detrick, Frederick, MD. LAITINEN, A. (1985). Ultrastructural organization of intraepithelial nerves in the human airway tract. Thorax 40,488-492.

LIPP, J., AND DOLA, T. (1980). Comparison of the efficacy of HS-6 versus HI-6 when combined with atropine, pyridostigmine and clonazepam for soman poisoning in the monkey. Arch. Int. Pharmacodyn. Ther. 246, 138-148. LIPPMANN. M., ALBERT, R. E., AND PETERSON. H. T.. JR. (197 I). The regional deposition of inhaled aerosols

AND SARIN

687

in man. In Inhaled Particles III. Vol. 1. Proceedings of an International Symposium Organized by the British Occupational Hygiene Society, London, 1970 (W. H. Walton, Ed.), pp. 105-122. Unwin Brothers Ltd., Old Woking. NIINIMAA, V., COLE, P., MINTZ, S.. AND SHEPHARD, R. J. (1980). The switching point from nasal to oronasal breathing. Resp. Physiol. 42,6 l-7 1. PENGELLY, L. D. (1977). Curve-fitting analysis of pressure-volume characteristics of the lung. J. Appl. Physiol. 42, 11 l-l 16. PESANTI, E. L. (1979). Kinetics of phagocytosis of Staphylococcus aureus by alveolar and peritoneal macrophages. Infect. Immun. 26,479-486. RICKETT. D. L., ADAMS, N. L., GALL, K. J., RANDOLPH, T. C., AND RYBCZYNSKI, S. (1982). Differentiation of Peripheral and Central Actions of Soman-Produced Respiratory Arrest, NTIS AD-A 117 29918. U.S. Army Medical Research Institute of Chemical Defense, Aberdeen Proving Ground, MD. SACKNER, M. A., GREENELTCH, D. M., HEIMAN, M. S., EPSTEIN, S.. AND ATKINS, N. (1975). Diffusing capacity, membrane diffusing capacity, capillary blood volume, pulmonary tissue volume, and cardiac output measured by a rebreathing technique. Amer. Rev. Resp. Dis. 111, 157-165. SAMPSON, S. R., AND VIDRUK, E. H. (1975). Properties of “irritant” receptors in canine lung. Resp. Physiol. 25,9-22.

STERRI, S. H.. ROGNERLJD,B., FISKUM, S. E., AND LYNGAAS, S. ( 1979). Effect of toxogonin and P2S on the toxicity of carbamates and organophosphorous compounds. Acta Pharmacol. Toxicol. 45,9- 15. TORJUSSEER,W. (1983). Nasal cancer in nickel. Histopathological findings and nickel concentrations in the nasal mucosa of nickel workers, and a short review of chromium and arsenic. In Nasal Tumors in Animals and Man. Vol. II. Tumor Pathology (G. Reznik and J. F. Stinson. Eds.), pp. 33-53. CRC, Boca Raton, FL. WIDDICOMBE, J. G. ( 1954). Respiratory reflexes from the trachea and bronchi of the cat. J. Physiol. (London) 123,55-70.

WINER, B. J. (197 1). Statistical Principles in Experimental Design, 2nd ed. McGraw-Hill, New York. WOLTHLJIS. 0. L., AND MEETER, E. (1968). Cardiac failure in the rat caused by diisopropyl lluorophosphate (DFP). Eur. J. Pharmacol. 2,387-392. ZAR, J. H. (1974). Biostatistical Analysis. Prentice-Hall, Englewood Cliffs, NJ.