Biological effects of short-term, high-concentration exposure to methyl isocyanate. III. Influence on gas exchange in the guinea pig lung.

The influence of methyl isocyanate (MIC) inhalation on the gas exchange function of the lungs in guinea pigs was studied by measuring arterial blood gases, pH, and tracheal pressure during constant-volume, artificial ventilation with air or 100% O2 at 40 and 120 min after exposure. A 15 min exposure to MIC at concentrations of 240 to 628 ppm caused a marked reduction in PaO2 and pHa and an elevated tracheal pressure during artificial ventilation. The low PaO2 was only slightly elevated when the animals were ventilated with 100% O2. Although the dry-wet lung weight ratio was reduced at the highest exposure concentration, the effect was not severe and no significant increase in lung water was found at the lower concentrations. MIC inhalation caused severe pulmonary blood shunting and ventilation/perfusion imbalance. This, in turn, led to hypoxemia, metabolic acidosis, and tissue hypoxia, which could produce death. The pulmonary gas exchange deficit likely resulted from bronchial and bronchiolar obstruction caused by sloughed epithelium and other debris from intra- and extrapulmonary airways.


Introduction
Following inhalation exposure to methyl isocyanate (MIC) at levels of 30 to 200 ppm, rats quickly exhibit respiratory distress with a slow, gasping breathing pattern (1,2). There is nasal and eye irritation, fluid discharge from the nose, and extensive damage to the epithelium in both upper and lower airways. These animals also exhibit a reduction in arterial pH (pHa) and partial pressure of oxygen (Pao2), and a rise in arterial partial pressure of carbon dioxide (Pao2), which is especially pronounced in animals that die during exposure.
The changes in arterial blood gases and pH may result because of damage to the respiratory control system, leading to hypoventilation, and/or because of gas exchange impairment in the lung. The current study was undertaken to determine if blood gases and pH were altered following inhalation of MIC, when the ventilation of the lung was held constant by use of a constantvolume respirator. We further attempted to identify the cause of rapid death in guinea pigs following exposure to high concentrations.

Animals and Animal Preparation
Seven adult, specific pathogen-free, Hartley, female guinea pigs (Hazleton Research Animals, Denver, PA), with an average body weight of442 + 51 g, were exposed to concentrations of MIC ranging from 240 to 628 ppm for 15 min in a static exposure chamber (3). Four control guinea pigs were studied without exposure to MIC. Immediately after removal from the exposure chamber, the animals were placed in a small enclosure containing 5% halothane in air. When the animals were in a surgical plane of anesthesia, their tracheae were cannulated and the animals were artificially ventilated with a constantvolume pump (Harvard Instruments, model 683) at a tidal volume of 3mL and a frequency of 76 breaths/min. These values were chosen to slightly hyperventilate the lungs of the control animals, compared to previous studies on unanesthetized, spontaneously breathing guinea pigs (4), with the goal of maintaining an optimal level of gas exchange in the anesthetized animals. A light sumgical plane of anesthesia was maintained by flowing air containing 1.5% halothane (Fluotec Mark III vaporizer) past the inlet of the respirator at a flow slightly higher than the peak inspiratory flow (Fig. 1). The gas was humidified at room temperature by bubbling through water. The left common carotid artery was cannulated (PE 50) for withdrawal of blood for blood gas and pH measurement. Body temperature was measured with a calibrated rectal thermistor inserted 6 cm into the lower gastrointestinal tract and a telethermometer (Yellow Springs Instrument, model 43). The signal from a second thermistor placed at the same depth was used by a temperature controller (Yellow Springs Instrument, model 72) to operate a heating pad to hold the body temperature near 37°C.

Measurements
Arterial blood gases and pH were measured using a Corning Model 170 blood gas analyzer that was calibrated with analyzed gases supplied by the manufacturer. Measurements were made at 370C but were corrected to the animal's body temperature when differences occurred. The pressure induced by inserting 3 mL of gas into the lungs was measured at the tracheal cannula with a pressure transducer (Gould, model P23 ID). The changes in pressure were recorded on a multichannel pen recorder (Grass, model 7D).
Blood samples were also taken for measurement of total white cells, erythrocytes, hematocrit, and hemoglobin on a Coulter Counter S-Plus IV. At the end of the experiment, the animals were killed with an overdose of sodium pentobarbital and the lungs were removed, examined for gross lesions, weighed, and dried for 72 to 96 hr at 50°C until further drying resulted in no additional reduction in weight. The dry/wet lung weight ratio was calculated for each animal.

Periodic Hyperinflation of the Lungs (Sighing)
In preliminary experiments on two control animals, it became apparent that gas exchange deteriorated after 2 hr of artificial ventilation (Table 1). There was a substantial reduction in Pao2 by this time. Subsequent animals were periodically hyperinflated by occluding the outflow of the pump for 3 to 4 tidal volumes twice at 20-min intervals and about 5 min prior to maldng measurements. This procedure, referred to as sighing, resulted in marked improvement of the Pao2 and resulted in no measurable deterioration of gas exchange function with time. Further, no areas of atelectasis were observed in the lungs of the control animals that were sighed, although such areas were observed in those that were not sighed. It appears that sighing is extremely important in maintaining normal gas exchange function in artificially ventilated guinea pigs.

Protocol
The animals were weighed and either exposed to MIC or directly anesthetized and studied. Following exposure to MIC, the guinea pigs were anesthetized and artificially ventilated with air and halothane. The first measurement of blood gases and other variables was made about 40 min after exposure and was repeated at 120 min after exposure. The air was then replaced by 100% oxygen, and the animals were ventilated for an additional 20 min. All measurements were then repeated. The animals were then killed and a necropsy was performed.

Data Analysis
A one-way analysis of variance was applied to data for each variable from the various treatment groups at each of the three measurement periods to determine if the various MIC treatments resulted in a significant change (p < 0.05). If a significant F value was obtained, a least significance difference test was used to determine which treatments differed from each other. All results are reported as means ± standard deviation (SD).

Results
Rapid, detrimental effects on gas exchange and pH were found at all MIC exposure concentrations used ( Table 2). Pao2 and pHa were severely decreased in all cases, even though a constant volume of gas entered the trachea. Although Pao2 increased with the increase in exposure level when the first measurements were made, it returned to nearly normal levels by 2 hr after exposure. Because the acidosis remained and the bicarbonate concentration dropped at the latter time, it is clear that a major component of the acidosis resulted from production of fixed acids and was, therefore, of metabolic origin. Pao2 and pHa did not appear to become worse with time after exposure and no animals died while being artificially ventilated.
To determine if the low Pao2 after MIC exposure was the result of intrapulmonary blood shunting or ventilation/perfusion inequality in the lung, we ventilated the animals for 20 min using 100% 02 and repeated the arterial blood gas and pH measurements. The expected high Pao2 was observed in guinea pigs not exposed to MIC, but dramatically reduced values occurred in all treated animals ( Table 3). The low Pao2 values suggest the likelihood of severe intrapulmonary blood shunting following MIC exposure but, because the Pao2 was somewhat improved over that during air-breathing, there was also some degree of ventilation/perfusion inequality.
Marked increases in tracheal pressure during constant volume ventilation following MIC exposure indicates likely reduction in the compliance of the lung (Fig.  2 and Table 4). Direct observation of the chest of these animals indicated less expansion in the exposed animals than in the controls.
Although a significant decrease in the lung dry/wet ratio (13% change) was found at the highest MIC exposure, the increase in water content was not dramatic (Table 5). There was no change in lung water content in the 240 ppm exposure group and only an 8.5% change in the 526 ppm group. These data suggest that even though the animals were exposed to high concentrations of MIC, the immediate disturbance in gas exchange was not caused by excess fluid in the lung.
During the course of the experiment, a small amount of blood was lost during surgery and approximately 2 mL of blood were taken for analyses. In addition, heparinized saline was injected to keep the catheter from clotting. To be certain that this procedure did not upset the hemodynamic balance of the animals, we measured several variables associated with 02 transport, namely hematocrit (Hct), hemoglobin (Hb), and red blood cell (RBC) count. Neither the experimental procedure nor   the MIC treatment caused an alteration in number of RBCs, Hb, or Hct. Previous studies on spontaneously breathing guinea pigs exposed to MIC suggest the occurrence of hemoconcentration (5). It is possible the added fluids in our experiment eliminated that effect. There was a significant elevation in white blood cell count in the highest MIC-treated group, but the cause is not known.
Gross observation of the lungs of the MIC-treated guinea pigs indicated atelectasis in areas of both right and left lungs. These areas were scattered throughout the lungs and did not seem to appear in any consistent regions in all animals. There were often multifocal hemorrhages in both lungs. Upon opening the trachea and primary bronchi, long strands of viscous material could be pulled out, often in a single piece with bifurcations extending down each primary bronchus. It is likely that this material occupied a large cross-sectional area of the lumen, especially of the smaller bronchi and bronchioles. The nasal passageways were also filled with a viscous, straw-colored material. None of the untreated animals had gross signs of lung abnormalities.

Discussion
These experiments demonstrated the severe and rapid influence of MIC inhalation on the gas exchange function of the lung. Inhaling this highly reactive compound resulted in extreme hypoxemia that undoubtedly led to severe tissue hypoxia and metabolic acidosis. The latter disturbances are the probable cause ofrapid death during or following high level exposure.
All evidence from this experiment suggests that the primary cause of hypoxemia following MIC inhalation is intrap.ulmonary blood shunting and ventilation/perfusion (V/Q) inequality. Measurement of Pao2 during inhalation of 100% 02 provides a test for lung units that receive no ventilation but are still perfused (intrapulmonary blood shunting). When 100% 02 is inhaled for several minutes, blood perfusing ventilated lung units equilibrates with the high alveolar Po2 but has an oxygen content only slightly higher than blood equilibrated by lung units previously ventilated with air. The blood 02 dissociation curve is nearly flat at a Po2 above 100 torr. When mixed venous blood passes through lung units that are not ventilated (shunted blood), no gas exchange occurs, and the Po2 and 02 content remain low. Mixing of the latter blood with that equilibrated with high alveolar Po2 results in a lowering of the 02 content of the mixture in proportion to the amount of shunted blood. Because of the nearly flat 02 dissociation curve at high Po2, a small reduction in 02 content leads to a large reduction in Po2 of the mixture that is easily detectable. The magnitude of the reduction in Po2 can be used to estimate the degree of the shunt. On the other hand, lung units that are poorly ventilated, relative to their perfusion (V/Q inequality), will also contain a high alveolar Po2 during the breathing of 100% 02, as in the case of lung units with a normal ventilation, provided enough time has been given for the washout of N2 present when air was breathed. Thus, mixed venous blood passing through these lung units will be equilibrated with a high alveolar Po2 and not cause a reduction in the Pao2 below that expected in a normal lung.
Because the Pao2 was not elevated substantially during 100% 02 breathing after exposure to MIC, some areas of the lung appeared to receive no ventilation but still had abundant perfusion; thus, intrapulmonary blood shunting was present. However, the Pao2 of the MIC-exposed animals did rise approximately 30 torr when they breathed 100% 02, suggesting that some lung areas were underventilated (but still receiving some ventilation) with respect to their perfusion.
The observed functional disturbances in gas exchange are consistent with bronchial or bronchiolar obstruction resulting from sloughed epithelial cells, secretions, and other debris, possibly derived from the airways proximal to the obstruction. Histopathological examination of the respiratory system following MIC exposure has demonstrated these plugs in both large and small airways (1,(6)(7)(8). Such an obstruction is also indicated by the high tracheal pressures that accompanied the 3-mL tidal volume delivered by the respirator and has been described in rats exposed to lower concentrations of MIC (9).
Arterial Pco2 was not excessively increased after MIC exposure in these pump-ventilated animals. This may result despite shunting and V/Q inequality because certain areas of the lung are likely to be overventilated by the constant tidal volume, thereby, removing excess amounts of CO2 from the blood perfusing these areas. The mixture ofblood from the overventilated areas with that from underventilated (or completely unventilated areas), therefore, might be relatively normal. The lack of evidence for extensive pulmonary edema suggests that the cause of shunting and V/Q imbalance was not alveolar flooding.
It is probable that the hypoxemia developed rapidly during MIC exposure, as indicated by the gasping breathing pattern of the animals in the chamber. That would quickly lead to tissue hypoxia and the resulting elaboration of lactic acid into the blood. The fixed acid would cause metabolic acidosis, resulting in a marked reduction in 02 affinity of hemoglobin and a further reduction in 02 transport to the tissues (10). If unchecked, this positive feedback would quickly lead to death, a common finding at exposure levels used in this study.
None of the artificially ventilated animals died during the course of this study. The tissue hypoxia and acidosis were still compatible with life, even though they were incurred in a very short time. The situation is likely to be compounded in spontaneously breathing animals because of possible impairment of the respiratory control system (1). Those animals very likely cannot maintain an adequate tidal volume because of the decreased compliance of the lung, irrespective of exaggerated ventilatory efforts. They then would be subject to hypoxemia and tissue hypoxia to an even greater extent than the artificially ventilated animals. Thus, prolonged survival may be possible after MIC exposure if assisted ventilation, oxygen therapy, and treatment of the metabolic acidosis are applied at an early enough time.