S. G. Kelsen

Effect of Elastase-induced Emphysema on the Force-generating Ability of the Diaphragm

The effect of emphysema on the ability of the diaphragm to generate force was examined in costal diaphragm muscle strips from 10 Golden hamsters killed 18 mo after intratracheal injection of pancreatic elastase in a dose producing hyperinflation (mean total lung capacity [TLC] = 163% of control) and generalized panacinar emphysema. 13 saline-injected normal animals served as controls. The time course of isometric tension and the effect of alterations in muscle fiber and sarcomere length on the isometric tension (T) generated in response to tetanizing electrical stimuli (length-tension [L-T] relationship) were examined. Elastase administration caused an increase in diaphragm muscle thickness and reduction in the length of costal diaphragm muscle fibers measured in situ. Emphysema significantly increased the maximum tetanic tension as a result of hypertrophy. Maximal tension corrected for increases in muscle cross-sectional area (T/cm(2)), however, was the same in emphysematous (E) and control (C) animals. Emphysema also shifted the muscle fiber L-T curve of the diaphragm but not of a control muscle, the soleus, toward shorter lengths. In contrast to the effects of E on the diaphragm muscle fiber L-T curve, the sarcomere L-T curve was the same in E and C. Since the length at which tension was maximal correlated closely with sarcomere number (r = 0.94; P < 0.001) reduction in the number of sarcomeres in series in muscles from emphysematous animals appeared to explain the shift in the muscle fiber L-T curve. We conclude that in elastase-induced emphysema adaptive changes both in diaphragm cross-sectional area and sarcomere number augment the force-generating ability of the diaphragm. We speculate that changes in sarcomere number compensate for alterations in muscle fiber length resulting from chronic hyperinflation of the thorax, while diaphragmatic muscle hypertrophy represents a response to changes in respiratory load and/or diaphragm configuration (LaPlace relationship).

THE ROLE OF AIRBORNE BACTERIA IN THE CONTAMINATION OF FINE PARTICLE NEBULIZERS AND THE DEVELOPMENT OF NOSOCOMIAL PNEUMONIA

Previous studies have shown that reservoir nebulizers are exceedingly vulnerable to bacterial contamination by gram negative organisms and produce aerosols containing large numbers of bacteria.,2 These contaminated aerosols have been shown to cause bacterial colonization of the respiratory tract and nosocomial pne~monia . .~ Several years ago, we studied the role of airborne bacteria in the contamination of fine particle nebulizers by comparing the incidence of contamination in groups of nebulizers placed in hospital locations whose airborne bacterial flora diff ered.4 More recently, we have examined the relationship between airborne bacteria and nosocomial attack rates in a surgical intensive care unit. In this unit, both the incidence of nosocomial respiratory tract infections and the bacterial profile of the ambient air were determined serially over a 5-year period. The data from these studies indicate that: 1) contamination of fine particle nebulizers can be produced by the airborne route; and 2) the incidence of nosocomial respiratory infections is significantly related to airborne bacterial levels.

Mechanisms underlying CO2 retention during flow-resistive loading in patients with chronic obstructive pulmonary disease.

The present study examined the respiratory responses involved in the maintenance of eucapnea during acute airway obstruction in 12 patients with chronic obstructive disease (COPD) and 3 age-matched normal subjects. Acute airway obstruction was produced by application of external flow-resistive loads (2.5 to 30 cm H2O/liter per s) throughout inspiration and expiration while subjects breathed 100% O2. Application of loads of increasing severity caused progressive increases in PCO2 in the patients, but the magnitude of the increase in PCO2 varied substantially between subjects. On a resistance of 10 cm H2O/liter per s, the highest load that could be tolerated by all COPD patients, the increase in PCO2 ranged from 1 to 11 mm Hg, while none of the normal subjects retained CO2. Based on the magnitude of the increase in PCO2 the patients could be divided into two groups: seven subjects whose PCO2 increased by less than or equal to 3 mm Hg (group I) and five subjects whose PCO2 increased by greater than 6 mm Hg (group II). Base-line ventilation and the pattern of breathing were similar in the two groups. During loading group I subjects maintained or increased tidal volume while all group II patients decreased tidal volume (VT). The smaller tidal volume in group II subjects was mainly the result of their shorter inspiratory time as the changes in mean inspiratory flow were similar in the two groups. The magnitude of CO2 retention during loading was inversely related to the magnitude of the change in VT (r = -0.91) and inspiratory time (Ti) (r = -0.87) but only weakly related to the change in ventilation (r = -0.53). The changes in PCO2, VT, and Ti during loading correlated with the subjects maximum static inspiratory pressure, which was significantly lower in group II as compared with group I patients. These results indicate that the tidal volume and respiratory timing responses to flow loads are impaired in some patients with COPD. This impairment, presumably due to poor inspiratory muscle function, appears to lead to CO2 retention during loaded breathing.

The effect of vagal blockade on the variability of ventilation in the awake dog

The present study examined the variability of breathing in five (5) awake tracheostomized dogs with the vagi intact and during complete vagal blockade produced by cooling exteriorized cervical vagal loops (VC). Breath by breath variations in both respiratory timing (assessed from the airflow signal) and the drive to the respiratory muscles (as assessed from the rate of inspiratory airflow (VI/TI) and occlusive pressure (P100) were examined. The degree of variability in the parameters characterizing breathing was evaluated from frequency distribution histograms and by calculation of the standard deviation. VC increased the mean values of VT, TI, TE, TI/TTOT, and decreased VT/TI and occlusion pressure, but had no consistent effect on the mean value of VE. The variability of VE, PACO2, VT, TI, TE, TI/TTOT was greater during VC in 4 of the 5 dogs. The increased variability of VE and PACO2 during VC appeared to be due to a poorer correlation between TI and TE. The present study suggests that vagal mechanoreceptors, presumably pulmonary stretch receptors, minimize breath by breath fluctuations in both the level and pattern of ventilation by controlling respiratory timing. An explanation, based on the model of inspiratory off-switching proposed by Beadley et al. (1975) is invoked.

Distribution of motor activity to expiratory muscles during sciatic nerve stimulation in the dog

The present study examined the effect of increasing sensory input from the lower limbs (assessed from the response to electrical stimulation of the sciatic nerve) on the distribution of electrical activity to the expiratory muscles. Expiratory muscle response to sciatic nerve stimulation (SNS) was compared to the response of the inspiratory muscles, and to their response to hypercapnia. In 16 anesthetized dogs afferent SNS increased ventilation and augmented the integrated EMG of all six expiratory muscles studied. Increases in abdominal muscle electrical activity were not uniform, being greater in the transversus abdominis and external oblique as compared to the internal oblique and the rectus abdominis. Increases in thoracic expiratory and inspiratory muscle activity during SNS were similar in magnitude. SNS performed while dogs were breathing 7% CO2, produced increased neural activity similar to those observed during O2 breathing. During CO2 rebreathing, at equal levels of minute ventilation, expiratory muscle responses to SNS and to CO2 were similar. In contrast, the rate of rise of the inspiratory muscle EMGs was greater during SNS. The present study indicates that the abdominal muscles participate in the respiratory response to afferent neural drive from skeletal muscles. The magnitude of their response is independent of their pre-stimulation level of activity and is similar to that observed during CO2 stimulated breathing.

Relationship between diaphragmatic activation and twitch tension to superimposed electrical stimulation in the cat

The purpose of these experiments was to evaluate the validity of the twitch-occlusion method as an index of the extent of diaphragmatic activation and to assess the extent of diaphragmatic activation during inspiration. Studies were performed in situ on innervated, perfused muscle strips from the costal region of the diaphragm in ten anesthetized cats. We measured isometric tension generated by the diaphragm during inspiration and following an interpolated electrical stimulation (2 Hz, 100 microseconds, 3.0 x threshold) of the nerve. The extent of MU activation was assessed by comparing twitch amplitudes during electrical stimulation applied in expiration and in inspiration. Spontaneous inspiratory activity was induced by adding CO2 to the inspired oxygen. Within an animal, the relationship between diaphragmatic activation and twitch occlusion was linear (range of r values was from -0.88 to -0.94). The extent of spontaneous diaphragmatic activation was normalized by dividing tension at end inspiration by the average twitch tension caused by stimuli applied during expiration. Across animals, twitch amplitude was inversely related to diaphragmatic activation (y = -0.36x + 1.13, r = -0.94). At a respiratory drive with end-tidal PCO2 approximately 1% above apneic threshold (end-tidal PCO2 between 5 and 6%), twitch occlusion was less than 5.0%. Increasing end-tidal PCO2 to at least 5% above apneic threshold (end-tidal PCO2 between 9 and 11%), twitch occlusion was still less than 50%. These results from a preparation that allows direct measurement of isometric tension of the diaphragm show that the interpolated twitch is linearly related to the extent of muscle activation through a broad range of muscle activity. In addition, these data indicate that a higher respiratory drives there exists a large reserve in the phrenic motor pool.

Effect of fatiguing resistive loads on the level and pattern of respiratory activity in awake goats

The effect of respiratory muscle fatigue on inspiratory muscle electrical activity (EMG), transdiaphragmatic pressure and ventilation during spontaneous breathing was examined in three awake goats. Studies were performed during progressive hypercapnia before and immediately after inspiratory muscle fatigue induced by flow resistive loading (IRL). IRL caused a decrease in the high-low ratio of the diaphragm and intercostal EMG and a decrease in Pdi during electrophrenic stimulation. After IRL, inspiratory time, the breathing duty cycle (inspiratory time/total breath cycle time), peak integrated activity of the diaphragm and external intercostal EMG per breath and per minute were all decreased at any given level of PCO2. Changes in the timing of respiratory motor activity and reduced muscle performance after IRL resulted in a decrease in transdiaphragmatic pressure and ventilation during hypercapnia. In conscious goats studied during spontaneous, chemically stimulated breathing, inspiratory muscle fatigue is associated with reductions in diaphragm and external intercostal muscle electrical activity and reductions in transdiaphragmatic pressure and ventilation.

Effect of Histamine on Respiratory Chemosensitivity in Conscious Goats

The effect of histamine on occlusion pressure and electrical activity of the diaphragm was studied in 5 conscious goats under conditions of changing respiratory drive by exposing the animals in separate trials to progressive hypercapnia produced by rebreathing technique. Both electrical activity of the diaphragm and occlusion pressure increased with hypercapnia. At any level of PCO2, occlusion pressure and diaphragm electrical activity were greater after histamine than in the control state at the same level of chemical stimulation. Both parameters changed proportionally, and a linear correlation was found between them (r greater than 0.9). These results in conscious unsedated animals indicate that (1) histamine causes an increase of inspiratory neuromuscular drive, and (2) occlusion pressure satisfactorily indicates changes in respiratory neuron motor output.

Inspiratory muscle activity during pulmonary edema in anesthetized dogs

Pulmonary edema is known to induce a rapid and shallow breathing pattern. However, its effects on the level and pattern of distribution of motor activity to the respiratory muscles is unclear. In the present study we evaluated the effect of oleic acid induced pulmonary edema on the electrical activity of the inspiratory muscles (costal and crural diaphragm and parasternal and external intercostal muscles) in the dog, and related it to the transdiaphragmatic pressure and ventilatory parameters over the course of CO2 rebreathing. Pulmonary edema, reflected by a 7.1 +/- 0.6 wet to dry ratio, decreased lung compliance by 57%, increased pulmonary shunt to 35%, and was associated with a rapid and shallow breathing pattern. When compared at equal levels of PCO2 during CO2 rebreathing before and during edema, ventilation and mean inspiratory flow were increased only at lower levels of hypercapnia and their responses to increasing levels of PCO2 were significantly diminished during edema. Transdiaphragmatic pressures were elevated during edema as compared to control values. The rate of rise of the electrical activity of all inspiratory muscles increased significantly during edema at all levels of PCO2. Peak activity, however, remained unchanged, due to shortening of the inspiratory duration. The EMG responses to progressive hypercapnia were not affected by edema. Pulmonary edema did not change the pattern of breathing and neural output to the inspiratory muscles in vagotomized dogs. We conclude that stimulation of pulmonary proprioreceptors during edema increases neural output to all inspiratory muscles. The neural response to hypercapnia is not altered by edema, and is additive to the vagal input. The ventilatory response to CO2 is blunted during severe edema, due to alterations in lung mechanics.