| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Laboratory of Cardiorespiratory Physiology, Brussels School of Medicine, 2Intensive Care Unit, Saint-Pierre University Hospital, and 3Chest Service, Erasme University Hospital, Brussels, Belgium
Submitted 20 July 2006 ; accepted in final form 2 November 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
respiratory muscles; diaphragm; intercostal muscles
Several attempts have been made to assess the effects of ascites on the function of the inspiratory muscle pump in humans and to evaluate the potential role of a pump dysfunction in causing dyspnea, but these studies have provided conflicting results. Specifically, Prezant et al. (29) reported an increase in pressure during maximum static inspiratory efforts in six patients with chronic renal failure receiving 3 liters of peritoneal dialysate. On the other hand, Siafakas et al. (32) reported that 2 liters of dialysate in similar patients caused a small reduction in pressure, and Duranti et al. (16) found no change in pressure after removal of 3.5–13.0 liters of ascites in eight patients with cirrhosis. However, in most patients, the amount of liquid in the peritoneal cavity was small and unlikely to impact significantly on the respiratory system. In addition, because the pressure measurements in these studies were obtained during voluntary efforts, they had a strong volitional component that could be affected by the nature and severity of the underlying disease process.
This volitional component was eliminated in the study by Hubmayr et al. (20). In their attempt to identify the determinants of the pressure-generating ability of the diaphragm, these investigators measured transdiaphragmatic pressure (Pdi) during isolated stimulation of the phrenic nerves in anesthetized dogs, first before and then after introduction of Ringer solution (60–100 ml/kg body wt) into the abdominal cavity. They found that the liquid introduction elicited an expansion of the lower rib cage, together with a cranial displacement and a small lengthening of the relaxed diaphragm. Pdi during phrenic nerve stimulation was also slightly increased. However, the volume of ascites in this study was equivalent to 4–7 liters in a 70-kg human subject. Therefore, although this volume was greater than that in patients on peritoneal dialysis (29, 32), it was still moderate compared with that in patients with cirrhosis or peritoneal metastases. More importantly, Hubmayr et al. did not report on the separate pressure changes on the pleural and abdominal side of the diaphragm. Consequently, as for the human studies previously discussed (16, 29, 32), it is difficult to relate the findings of this study to dyspnea.
The present study was initially undertaken to test the hypothesis that severe ascites increases the impedance on the diaphragm and, thereby, impairs the lung-expanding action of the muscle. Thus progressively increasing ascites was induced in dogs, and the changes in pleural pressure and abdominal pressure (Pab), as well as the displacements of the diaphragm, were examined during isolated stimulation of the phrenic nerves in the neck. The results confirmed that severe ascites has a marked detrimental effect on the lung-expanding action of the diaphragm, and this prompted us subsequently to evaluate how the respiratory muscle pump overall adapts to this impairment and performs the act of breathing.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
15- to 20-mm intervals, and those attached to the caudal half, in the zone of apposition of the diaphragm to the rib cage (27), were spaced at 25- to 30-mm intervals. Consequently, the chord length between the successive markers closely approximated the arc length along the diaphragm. The abdomen was finally sutured in two layers, and two experimental protocols were followed.
Experiment 1.
Six animals were studied first to evaluate the effects of ascites on the elastance of the abdomen, the pressure-generating ability of the diaphragm, and the silhouette and length of the muscle during isolated contraction. In each animal, the C5 and C6 phrenic nerve roots were isolated bilaterally in the neck and laid over two pairs of insulated stainless steel stimulating electrodes, and a differential pressure transducer (Validyne, Northridge, CA) was connected to a side port of the endotracheal tube to measure the change in airway opening pressure (
Pao). A long inextensible thread was also attached to the abdominal suture and led ventrally, perpendicular to the coronal plane, over a pulley placed above the animal, and it was connected to a linear displacement transducer (Schaevitz Engineering, Pennsauken, NJ) to measure the changes in the anteroposterior (AP) diameter of the abdomen.
The animal was allowed to recover for 15 min after instrumentation, after which it was made apneic by mechanical hyperventilation. Abdominal elastance was then assessed by measuring the changes in Pab and AP diameter of the abdomen during passive inflation of the respiratory system. Five levels of inflation (200, 400, 600, 800, and 1,000 ml) were applied in duplicate. When this procedure was completed, the animal was reconnected to the ventilator and hyperventilated. After the ventilator was stopped, the endotracheal tube was occluded at resting end expiration [functional residual capacity (FRC)], and the changes in Pao and Pab generated by bilateral, tetanic stimulation of the C5 and C6 phrenic nerve roots were measured; the stimuli were square pulses of supramaximal voltage, 0.1-ms duration and 20-Hz frequency. Two trials of stimulation were performed. AP radiographs of the lower rib cage and upper abdomen were also taken in the four animals with markers along the diaphragm, first during relaxation at FRC and then during stimulation of the phrenic nerves.
After completion of these measurements, a volume of liquid corresponding to 25 ml/kg body wt was introduced into the abdominal cavity, and measurements of abdominal elastance and pressure changes produced by phrenic nerve stimulation were repeated. Liquid volume was subsequently increased by increments of 25 ml/kg body wt to a total of 200 ml/kg, and measurements of abdominal elastance and pressure changes during phrenic nerve stimulation were obtained after each infusion. Chest radiographs during relaxation and during phrenic nerve stimulation were also taken at 50, 100, 150, and 200 ml/kg. The liquid introduced into the abdominal cavity (Dianeal glucose 1.36%, Baxter) was isosmotic, so its volume after injection remained stable.
In each animal, liquid was finally removed from the abdominal cavity, such that the total volume of ascites was brought back to 100 ml/kg, and the changes in Pao and Pab during phrenic nerve stimulation were measured again. The pressure changes in this condition were only 8.2 ± 3.7% (range –10.0 to +10.0%) different from those obtained during the initial measurement, thus indicating that the phrenic nerve preparation was stable.
Experiment 2. Eight animals were next studied to assess the response of the respiratory muscle pump to the diaphragmatic dysfunction induced by ascites (see RESULTS). The phrenic nerve roots in these animals were left intact, but the vagi were isolated in the neck and sectioned to avoid the tachypnea that ascites, acting through the decrease in FRC and the increase in vagal afferent inputs from the lung, would probably elicit otherwise. Also, the rib cage and intercostal muscles were exposed on the right side of the chest from the second to the ninth rib by deflection of the skin and underlying muscle layers, and a pair of silver hook electrodes spaced 3–4 mm apart was inserted into the third interspace to record the electromyogram (EMG) of the parasternal intercostal muscle. The electrodes were inserted in parallel fibers in the muscle bundles situated near the sternum, i.e., in the muscle area with the greatest inspiratory drive (15, 23). The EMG signal was processed with an amplifier (model 830/1, CWE, Ardmore, PA), band-pass filtered below 100 and above 2,000 Hz, and rectified before its passage through a leaky integrator with a time constant of 0.2 s. In each animal, a hook was also screwed into the fourth rib in the midaxillary line and connected to a linear displacement transducer (Schaevitz Engineering) to measure the craniocaudal (axial) displacement of the ribs during breathing, and a heated Fleisch pneumotachograph coupled with a differential pressure transducer was connected to the endotracheal tube to measure the changes in lung volume.
After 30 min of recovery, baseline measurements of tidal volume, Pab, rib motion, and parasternal EMG activity during resting breathing were obtained. After two runs of 15 breaths were recorded over a 20-min period, liquid was introduced into the abdominal cavity. Liquid volume was increased by increments of 50 ml/kg to a total of 200 ml/kg. As during control, two runs of resting breathing were obtained in each condition. Earlier studies clearly established that prolonged anesthesia in dogs causes the gradual development of atelectasis and a decrease in pulmonary compliance (9, 27). To eliminate the confounding influence of this factor on the response to ascites, the animal in each condition was given a series of 10 large passive inflations before the measurements were obtained.
The animals in experiment 1 were maintained at a constant, rather deep, level of anesthesia throughout the observation. They had no corneal reflex, no movements of the fore- or hindlimbs, and no changes in heart rate and blood pressure, including during phrenic nerve stimulation. In contrast, to avoid obscuring the response studied, in experiment 2, we regulated anesthesia such that the animals’ blink and corneal reflexes were present throughout the measurements. Rectal temperature in these animals was also maintained constant between 36 and 38°C with infrared lamps. At the end of the experiment, the animal was given an overdose of anesthetic (30–40 mg/kg iv).
Data analysis. For each volume of ascites in each animal of experiment 1, the increases in Pab and AP diameter of the abdomen during passive inflation were averaged over the two trials, and the slope of the relation between the two variables was calculated by linear regression techniques (coefficient of correlation = 0.952–0.999) to yield abdominal elastance. The pressure changes obtained during phrenic nerve stimulation were also averaged over the two trials, and the displacements of the diaphragm were assessed in two ways, as previously described (13, 14). 1) The contours of the muscle were traced during relaxation and during stimulation; all contours were related to a metallic marker that was attached to the sling on the side of the animal and was, therefore, stationary. For each volume of ascites, the contour during phrenic nerve stimulation was then superimposed on that during relaxation, and the axial displacement of the diaphragmatic dome in the two sagittal planes situated midway between the spinous processes of the vertebrae and the lateral rib cage margins was measured. The perpendicular distance from the sagittal midplane to the lateral margin of the lower rib cage during stimulation was also measured to provide an estimate of the radius of the ring of insertion of the diaphragm into the lower rib cage. 2) Diaphragmatic muscle length in each condition was quantified by measuring the linear distance between adjacent radiopaque markers and by summing the distances between markers in each row. To allow comparison between the different animals, however, the changes in muscle length during stimulation were expressed as percent changes relative to the muscle length during relaxation (Lr). Because phrenic nerve stimulation produced identical length changes in the right and left hemidiaphragms and identical displacements of the domes, these changes were averaged for each individual animal.
For each volume of ascites in each animal of experiment 2, tidal volume, Pab, the inspiratory axial displacement of the rib, and phasic inspiratory EMG activity in the parasternal intercostal muscle were averaged over 10 consecutive breaths from each run. Inspiratory EMG activity was first quantified by measuring the peak height of the integrated EMG signal in arbitrary units, and it was then expressed as a percentage of the activity recorded before ascites (control). In addition, inspiratory time (TI) was measured from the onset of the parasternal intercostal EMG until peak activity, and the peak deflection of the integrated EMG signal was also divided by TI to obtain the average rate of rise of inspiratory activity.
Data were finally averaged across the animal group, and they are presented as means ± SE. ANOVA with repeated measures was used for statistical analysis of the effects of increasing ascites on abdominal elastance, pressure, diaphragm displacement, and diaphragm muscle length (experiment 1), and Student-Newman-Keuls tests were used, when appropriate, for multiple comparison testing of the mean values. Statistical assessment of the effects of ascites on tidal volume, Pab, axial rib displacement, and parasternal intercostal EMG activity during spontaneous breathing (experiment 2) was made similarly. The criterion for statistical significance was taken as P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Pao and Pab during phrenic nerve stimulation in the different conditions are shown in Fig. 2. Pao remained unchanged until ascites was 50 ml/kg, and it then decreased progressively and continuously as ascites was increased from 75 to 200 ml/kg (P < 0.001). At 200 ml/kg, therefore, Pao was only 25.5 ± 3.3% of the control value. In contrast, Pab increased with increasing ascites up to 125 ml/kg (P < 0.001) and then decreased as ascites was increased further. As a result, Pdi (Pab – Pao) remained unchanged until ascites was 100 ml/kg and decreased progressively from 59.8 ± 5.0 to 33.3 ± 4.7 cmH2O as ascites was increased from 100 to 200 ml/kg (P < 0.001).
|
|
|
|
|
Pab increased (P < 0.001) from 2.86 ± 0.30 cmH2O during control to 4.86 ± 0.80 cmH2O at 150 ml/kg (Fig. 7C). As ascites was increased to 200 ml/kg, however, Pab decreased slightly to 4.46 ± 0.79 cmH2O.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Pao during isolated contraction of the diaphragm would be related to the effective pressure (or force) exerted by the muscle (PA) and to the elastance of the rib cage (KR) and abdominal (KA) compartments, such that
 | (1) |
Pao would be smaller for a given PA.
In agreement with our hypothesis, every animal of the study showed a progressive increase in abdominal elastance as the volume of ascites was increased (Fig. 1). Every animal also showed a gradual decrease in Pao during isolated phrenic nerve stimulation (Fig. 2), thus confirming that ascites does adversely affect the lung-expanding action of the diaphragm. The values of Pdi obtained in these animals suggest, however, that a factor other than the increase in abdominal elastance operates to impair diaphragmatic function in this condition. Thus the relaxed diaphragm in supine dogs is near its optimum force-producing length (17, 30), and, with ascites, its muscle fibers lengthened 3–4% (Fig. 3). To the extent that the muscle fibers still shortened by 11% during phrenic nerve stimulation with severe ascites, the conclusion can therefore be drawn that these fibers continued to operate on the ascending portion of their length-tension characteristics (17, 25, 30). Furthermore, the finding that the amount of diaphragm shortening during stimulation decreased with ascites (Fig. 3) implies that the diaphragmatic fibers during contraction were longer. Consequently, the tension developed by the muscle in response to a given activation should be greater; hence, one would expect that Pdi would be greater and that the adverse effect of abdominal elastance on Pao would be partly offset. In fact, although the increase in abdominal elastance with moderate ascites induced an increase in Pab during phrenic nerve stimulation, this increase was relatively small and hardly compensated for the loss in Pao. As a result, Pdi remained essentially unchanged. When ascites was severe, Pab even decreased, such that Pdi decreased as well.
Boriek et al. (6) examined the determinants of Pdi in dogs. Measuring the changes in diaphragm length and diaphragm curvature as well as Pdi, these investigators showed that curvature remained virtually constant during spontaneous inspiratory efforts at different lung volumes. They concluded, therefore, that Pdi during such efforts was exclusively related to muscle length. However, they also predicted that an expansion of the lower rib cage would induce a decrease in diaphragm curvature and, with it, a decrease in the pressure developed. Such a lower rib cage expansion has previously been shown to occur with ascites (20), and it was also seen in the present study. Thus, when the volume of ascites in our animals was set at 200 ml/kg, the radius of the ring of insertion of the diaphragm to the lower rib cage was 23% greater than during control (Fig. 4), and the analysis of Boriek et al. would predict that, for a given muscle tension, such an increase would lead to a 25% reduction in Pdi. Although the decrease in Pdi for this volume of ascites exceeded the predicted value by a factor of 2, this result supports the idea that, in severe ascites, the loss in the lung-expanding action of the diaphragm is, in part, the result of the lower rib cage expansion.
To the extent that the diaphragm is the main inspiratory muscle, it would be expected that a marked decrease in its lung-expanding action, as observed during severe ascites, would have a detrimental effect on lung expansion during breathing. Indeed, as ascites was progressively increased, tidal volume and minute ventilation decreased gradually, leading to a prominent increase in arterial PCO2 (Figs. 5 and 6). However, arterial PO2 remained essentially unchanged until ascites was 200 ml/kg. This confirms that the passive inflations performed in our animals were effective in preventing pulmonary atelectasis and, therefore, that the decrease in tidal volume was primarily related to the decrease in the lung-expanding action of the diaphragm, rather than a decrease in pulmonary compliance.
The result of this increased hypercapnic drive was an increase in parasternal intercostal inspiratory activity and, with it, an increase in the inspiratory elevation of the ribs (Figs. 5 and 7). Thus, apparently, the response of the canine respiratory muscle pump to severe ascites is qualitatively similar to its response to diaphragmatic paralysis (8). The diaphragm in our animals, however, was not paralyzed. No attempt was made to record diaphragmatic EMG activity in the present study, because this activity, when recorded with intramuscular electrodes, is well known to show artifactual changes with alterations in the conductivity of the environment surrounding the electrodes (7) or in the length of the muscle fibers (19, 22). However, similar to neural drive to the parasternal intercostals (10, 11), neural drive to the diaphragm is primarily governed by supraspinal control mechanisms, and measurements of motoneuron synchronization in cats by Vaughan and Kirkwood (34) showed that phrenic motoneurons and parasternal intercostal motoneurons receive common monosynaptic inputs. In addition, whereas an isolated increase in parasternal intercostal inspiratory activity would produce a decrease in Pab during inspiration, Pab during inspiration increased with ascites (Fig. 7C). This increase amounted to 70% of the control value; i.e., its magnitude was similar to that observed during isolated stimulation of the phrenic nerves (Fig. 2). It is most likely, therefore, that ascites induced a compensatory increase in neural drive to both the diaphragm and the parasternal intercostals, and the result of this compensation was that the fall in tidal volume, although prominent, was smaller than that anticipated on the basis of the loss of Pao during isolated phrenic nerve stimulation. That is, with an ascites of 200 ml/kg, the fall in Pao (and pleural pressure) during isolated phrenic stimulation was 75% of the control value (Fig. 2), whereas during spontaneous breathing the fall in tidal volume was 48%.
The model of ascites investigated in this study differs from the situation in patients with liver or peritoneal diseases in several respects. 1) The animals were anesthetized, and vagal afferent inputs were suppressed. 2) Pulmonary atelectasis in base of the lung was prevented by repeated lung inflation. 3) Ascites in the animals was produced acutely, whereas in patients it develops slowly over several weeks or months and might, therefore, induce remodeling in the diaphragm and abdominal muscles. Studies on limb muscles in cats and mice (33, 35) and on the diaphragm in hamsters (18, 21) have shown that chronic muscle shortening causes a loss of sarcomeres in series along the muscle fibers. Conversely, when limb muscles in cats and mice are immobilized for a few weeks in a lengthened position, sarcomeres are added (33, 35), and the result is that the length of individual sarcomeres is virtually restored to its initial value. Therefore, it is possible that, in patients with ascites, the increase in passive diaphragmatic and abdominal muscle tension and, with it, the increase in abdominal elastance are partly offset by an addition of sarcomeres. Nonetheless, according to the conventional current theory, dyspnea arises from a mismatch between central respiratory motor activity and incoming afferent information from the chest wall or pulmonary receptors (2, 24). By demonstrating that, in the dog, ascites impairs the lung-expanding action of the diaphragm and induces a compensatory, but insufficient, increase in neural drive to the inspiratory muscles, the present study may account, at least in part, for dyspnea in this condition.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. Leduc and A. De Troyer Impact of acute ascites on the action of the canine abdominal muscles J Appl Physiol, June 1, 2008; 104(6): 1568 - 1573. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. van Lunteren Beer and the respiratory muscles: the adverse effects of ascites J Appl Physiol, June 1, 2008; 104(6): 1551 - 1552. [Full Text] [PDF]
|
|
|
|
|
|
D. Leduc, M. Cappello, P. A. Gevenois, and A. De Troyer Mechanics of the canine diaphragm in ascites: a CT study J Appl Physiol, February 1, 2008; 104(2): 423 - 428. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. D. Troyer and D. Leduc Role of pleural pressure in the coupling between the intercostal muscles and the ribs J Appl Physiol, June 1, 2007; 102(6): 2332 - 2337. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
