Contrary to adaptive support ventilation (ASV), prolonged totally controlled mechanical ventilation (CMV) results in the absence of diaphragm activity and causes ventilator-induced diaphragmatic dysfunction. Because maintaining respiratory muscles at rest is likely a major cause of ventilator-induced diaphragmatic dysfunction, ASV may prevent its occurrence in comparison with CMV. The aim of our study was to compare the effects of ASV with those of CMV on both in vivo and in vitro diaphragmatic properties.
Two groups of six anesthetized piglets were ventilated during a 72-h period. Piglets in the CMV group (n = 6) were ventilated without spontaneous ventilation, and piglets in the ASV group (n = 6) were ventilated with spontaneous breaths. Transdiaphragmatic pressure was measured after bilateral, supramaximal transjugular stimulation of the two phrenic nerves. A pressure-frequency curve was drawn after stimulation from 20 to 120 Hz of the phrenic nerves. Diaphragm fiber proportions and mean sectional area were evaluated.
After 72 h of ventilation, transdiaphragmatic pressure decreased by 30% of its baseline value in the CMV group, whereas it did not decrease in the ASV group. Although CMV was associated with an atrophy of the diaphragm (evaluated by mean cross-sectional area of both the slow and fast myosin chains), atrophy was not detected in the ASV group.
Maintaining diaphragmatic contractile activity by using the ASV mode may protect the diaphragm against the deleterious effect of prolonged CMV, as demonstrated both in vitro and in vivo, in healthy piglets.
What We Already Know about This Topic
❖ Controlled mechanical ventilation leads to diaphragmatic dysfunction and atrophy
❖ Adaptive support ventilation in which patient effort is included to trigger breathing hastens respiratory weaning in postoperative patients
What This Article Tells Us That Is New
❖ In piglets, phrenic nerve-stimulated diaphragmatic strength was greater and histologic atrophy was less if they were ventilated for 72 h with adaptive support rather than controlled mechanical ventilation.
PROLONGED totally controlled mechanical ventilation (CMV) results in the complete absence of neural activation and mechanical activity of the diaphragm and has been shown to induce muscle atrophy, proteolysis, and reactive oxygen species liberation, leading to rapid losses in diaphragmatic function, a syndrome known as ventilator-induced diaphragmatic dysfunction (VIDD).1–4Few countermeasures to prevent VIDD have been evaluated. Continuous positive airway pressure,3intermittent spontaneous breathing,1or assist-control mechanical ventilation4has been shown to protect the diaphragm against the deleterious effects of CMV in animals. Our group demonstrated that maintaining spontaneous breathing activity with pressure support ventilation (PSV) reduced mechanical ventilation-induced proteolysis and inhibition of protein synthesis in comparison with CMV in an in vitro rat model.5Adaptive support ventilation (ASV), a complex mode recently approved by the Food and Drug Administration, is a new automatic ventilation mode, allowing assisted ventilation, in which minute ventilation is controlled by a combination of tidal volume (VT) and respiratory rate (RR) based on respiratory mechanics. In spontaneously breathing patients who are able to trigger a breath, the ventilator generates pressure support breaths, automatically adjusting inspiratory pressure to achieve the target VT, and it delivers additional pressure-controlled breaths if the RR of the patient is less than the target RR.6–8It has recently been shown that ASV may accelerate respiratory weaning after cardiac surgery in comparison with pressure-regulated, volume-controlled ventilation with an automode,9synchronized intermittent ventilation,10and pressure-controlled or pressure-support ventilation.11It has also been reported that ASV is feasible in the more severely ill patients of an intensive care unit12and reduces the work of breathing in comparison with PSV when a dead space is added to the ventilator circuit (i.e ., increase ventilatory demand) of critically ill patients.13
Therefore, we developed a model to examine whether ASV may protect the diaphragm against the detrimental effects of CMV, both in vivo and in vitro , in the same animal. We hypothesized that ASV would induce, in a healthy piglet model, less diaphragmatic dysfunction than CMV after 72 h of mechanical ventilation.
Materials and Methods
This study, including care of the animals involved, was conducted according to the official edict presented by the French Ministry of Agriculture (Paris, France) and the recommendations of the Helsinki Declaration. Therefore, these experiments were conducted in an authorized laboratory and under the supervision of authorized researchers (S.J., X.C., and S.M.).
We used the same experimental design described in our previous studies (fig. 1).14,15In brief, 14 piglets (15–20 kg) were anesthetized with intravenous pentobarbital sodium (5–6 mg/kg), intubated with a cuffed endotracheal tube, and mechanically ventilated (Galileo®; Hamilton Medical AG, Rhazuns, Switzerland), with an inspired fraction of oxygen of 0.35, a VTbetween 10 and 12 ml/kg, an RR from 15 to 30 cycles/min to obtain normocapnia, and 5 cm H2O of positive end-expiratory pressure. Neuromuscular blocking agents were not used. The piglets were anesthetized with continuous intravenous propofol (15–20 mg · kg−1· h−1), midazolam (0.1–0.3 mg/h), and ketamine (3–4 mg · kg−1· h−1). The level of sedation was monitored with bispectral index (BIS®; Aspect, Norwood, MA).16An oral gastric tube was placed. A vesicostomy was performed, and a urine catheter was placed for urine collection. A rectal probe was used for frequent temperature measurements, and the animals were rested on soft cushions. Heating pads were used as needed to maintain a normal body temperature of 38.5°–39.5°C. A carotidal arterial catheter (PiCCO®; Pulsion, Munich, Germany) was inserted for the monitoring of heart rate, arterial blood pressure, and cardiac output. Arterial pressure of carbon dioxide levels was checked by using a capnograph (Deltatrac®; Datex-Ohmeda, Helsinki, Finland) and then verified by arterial blood gases (iSTAT®; Abbott, Abbott Park, IL). Parenteral nutrition was given from the first day (10% glucose solution and 20% amino acids solution, and Hyperamine 20%®; Braun, Boulogne Billancourt, France) providing 30–35 kcal · kg−1· day−1. All procedures were performed aseptically. The animals received prophylactic intravenous antibiotics three times daily (amoxicillin–clavulanate, 100 mg · kg−1· day−1). At the end of the procedure, costal region diaphragm samples were removed and then animals were killed by intravenous injection of pentobarbital sodium and potassium chloride. A physician provided round-the-clock supervision and animal care for the entire duration of the study. The two groups received the same care, except for the ventilatory mode.
Seven piglets were ventilated for 72 consecutive hours in a totally controlled (CMV group) mode, and seven piglets were ventilated for 72 consecutive hours in a partially spontaneous mode (ASV group; fig. 2). In the CMV group, ventilatory settings were as follows: VTat 10–12 ml/kg of the ideal body weight, RR from 15 to 30 min−1, and positive end-expiratory pressure level at 5 cm H2O. The absence of spontaneous breathing was verified on the ventilator trend graphs, and electromyographic activity of the diaphragm was measured in a few animals to ensure that no electrical activity was present in the diaphragm. In the ASV group, ventilatory settings were pediatric mode setting in phase with ideal body weight of the piglet, inspiratory flow trigger at 0.3 l/min, percentage of mechanical ventilation between 100 and 150%, positive end-expiratory pressure level at 5 cm H2O, and expiratory trigger at 25% of peak inspiratory flow. Active piglets were allowed to breathe spontaneously, and the ratio of pressure-assisted cycles was verified on the ventilator trend graphs. In both groups, oxygenation was maintained with Fio2from 25 to 35% and minute ventilation to assess normocapnia. The deepness of anesthesia was adjusted to blow out the respiratory drive in the CMV group and to allow spontaneous breathing without awakening in the ASV group.
Assessment of Diaphragm Muscle Activity during Mechanical Ventilation
First Part of the Study: In Vivo Assessment of Transdiaphragmatic Pressure.
We measured transdiaphragmatic pressure (Pdi) for every 12 h to assess in vivo diaphragmatic contractile force in both groups as described in previous studies.14,15In brief, double air-filled balloon-tipped catheters were placed transorally into the distal third of the esophagus and in the stomach for measurement of Pdi. Bipolar transvenous pacing catheters were introduced via each internal jugular vein and adjusted to achieve stimulation of the phrenic nerve and subsequent contraction of the diaphragm. Pdi was produced by supramaximal stimulation at frequencies of 20, 40, 60, 80, 100, and 120 Hz in a serial manner. Each train of impulses lasted for 2,000 ms, and each pulse had duration of 150 ms. A pressure–frequency curve was obtained for both groups at each 12-h period and then compared.
Second Part of the Study: In Vitro Assessment of Diaphragm Histology.
Biopsies (1 cm3) from the apposition zone of the costal diaphragm from the entire midcostal muscle spanning were removed just before euthanasia. Each biopsy was partitioned, frozen in liquid nitrogen-cooled isopentane after 3–5 min for length equilibrium, and stored at −80°C. Immunochemistry was assessed on 10-μm unfixed serial transverse cryostat sections according to a well-described procedure.10In brief, histologic analysis was performed on six cross-sections from each muscle. Stained sections were visualized under a Nikon optiphot-2 microscope (Nikon Instruments Europe, B.V. Amstelveen, The Netherlands). All diaphragm images were obtained under identical conditions and with the same objective lens. The shape of each muscle fiber was accurately defined (on images ×400), and a schematic drawing of each stained diaphragm section was recorded. Schematic representations of each muscle section were analyzed by Histolab program (version 5-13-1, Microvision Instrument, license number; 2497, Evry, France), and data were averaged per approximately 500 muscle fibers from the dissected diaphragms of piglets from each group.
Ten-micrometer unfixed cryostat sections of CMV and ASV group diaphragms were stained by hematoxylin and eosin. On each schematic drawing, the fiber cross-sectional areas were measured. To assess relative fiber percentages, immunochemistry was assessed on 10-μm unfixed serial transverse cryostat sections. Two adjacent sections were incubated for 1 h in 1% bovine serum albumin with the respective primary antibodies for both the mouth anti–slow myosin heavy chain (M-8421; Sigma, Saint Louis, MO; dilution 1/8,000) and the mouth anti–fast myosin heavy chain (M-4276; Sigma; dilution 1/8,000). In addition, polyclonal antibodies against dystrophin (H4; dilution 1/5,000), produced and characterized in our laboratory, were included in each of these stains to visualize fiber membranes. Immunoreactions were detected with Cy3 and fluorescein isothiocyanate-conjugated goat anti-piglet. After washing with phosphate-buffered saline solution, sections were incubated 1 h at room temperature with Cy3-goat anti-piglet IgG (Chemicon International, Molsheim, France; dilution 1/5,000) to detect myosin heavy chain, and with fluorescein isothiocyanate (Chemicon International, AP 132F; dilution 1/1,000) to detect dystrophin. Unbound antibodies were removed by washing the sections in phosphate-buffered saline. For negative controls, only the second antibody was applied. Sections were then visualized under adapted fluorescence. By using these stains, the myosin heavy chains that react with the chosen antibody on each section appeared orange, whereas the dystrophin at the membrane appeared green. We compared fiber type proportion measurements between the groups.
Values are means and SD or medians and quartiles (25–75th), as required. Normality of the distribution was assessed with the Kolmogorov–Smirnov test. A two-way analysis of variance with time (H0, H12, H24, . . ., H72) as one factor and modality (CMV vs . ASV) as the other factor was used. When appropriate, a Newman–Keuls test was used. Nonparametric paired Wilcoxon tests were used to compare data from days 1 and 3 for each animal in both the CMV and ASV groups. All P values were two tailed, and a P value less than 0.05 was considered significant (StatView®, version 5.0; SAS Institute Inc., Berkeley, CA).
Systemic and Biologic Response to Mechanical Ventilation
Among the 14 piglets, two died before the end of the study: one because of myocardial infarction at H10 (CMV group) and one because of septic shock (peritonitis after bladder catheterization resulting in small bowel perforation in the ASV group) after 28 h of ventilation. These two piglets were subsequently excluded from the analyses.
No significant differences were observed between the CMV group and the ASV group for all the studied baseline variables. Long-term mechanical ventilation, either in CMV or ASV, did not have consequences on body weight, intestinal transit, diuresis (data not shown), or hemodynamic variables (table 1). BIS values were significantly different between the CMV and ASV groups (72 ± 8 vs . 42 ± 12, P = 0.02). Mean midazolam level administration remained at a higher level in the CMV group (3.8 ± 0.7 mg/h) than that in the ASV group (1.5 ± 0.7 mg/h) during the study (P < 0.05 between CMV vs . ASV after 12 h of ventilation). Doses of propofol did not significantly differ between the groups during the study (102 ± 8 and 90 ± 10 mg/h in the CMV and ASV groups, respectively).
Ventilatory Parameters and Spontaneous Breathing Data
In the CMV group, spontaneous activity of the diaphragm was present for no more than 5% of the delivered breaths, as determined by analysis of the trend graphs and the Pdi tracings. In the ASV group, active piglets triggered 80% of the delivered breaths, as assessed by the Galileo® trend graphs of ventilator and, for a few animals, the surface electromyogram activity (fig. 2). Once normal gas exchanges were obtained, ventilatory settings were not modified throughout the study (table 1).
In Vivo Assessment of Diaphragmatic Force
The baseline pressure–frequency curves of the two groups were not significantly different (Pdi = 23.4 ± 2.6 vs . 24.6 ± 4.8 cm H2O at 20 Hz and 45.8 ± 3.0 vs . 46.8 ± 1.5 cm H2O at 100 Hz for the CMV and ASV groups, respectively; fig. 3, A and B).
Furthermore, the decrease in the Pdi between H0 and H72 in the CMV group (−26% [−17 to −32]) was significantly higher than that in the ASV group (−2% [−8 to +3]; P < 0.05) at all frequencies of stimulation, except at 40 and 60 Hz at which the difference did not reach the statistical significance (fig. 3). Figure 4shows the evolution of Pdi over time at a stimulation frequency of 100 Hz. In the CMV group, Pdi decreased significantly after 48 h of ventilation (P < 0.05). In the ASV group, Pdi did not significantly differ during the whole study period.
In Vitro Assessment of Diaphragmatic Histology
Figure 5Ashows photomicrographs of typical diaphragm muscle cross-sections obtained in one animal of each group. Fibers that react with the antibody appear orange in figure 5B, whereas fibers not reacting with the antibody appear black. A representative slow-twitch fiber is indicated by an open square and a fast-twitch fiber by an open circle. In the CMV group, we observed a marked atrophy of both slow- and fast-twitch fibers after 72 h of totally controlled ventilation. Indeed, after 72 h of CMV, fiber cross-sectional area was decreased by 30–40% for both slow- and fast-twitch fibers (P < 0.05; fig. 5C). The atrophy reported in the CMV group contrasts with that reported in the ASV group, which was not associated with fiber cross-sectional area variation after 72 consecutive hours of ventilation (fig. 5C). Atrophy in the CMV group concerned both slow- and fast-twitch fibers, and fiber proportion at H72 was not different in comparison with baseline in both groups (fig. 5D).
This study demonstrates that maintaining spontaneous ventilation with ASV is efficient to protect the diaphragm against the occurrence of VIDD, both in vivo and in vitro , in the same piglet model. In addition, we report that in vivo , both diaphragmatic contractility and diaphragmatic atrophy occur when a totally controlled ventilation mode is applied but are prevented when spontaneous cycles with ASV are maintained.
By using in vitro animal models, several studies have reported that CMV-induced diaphragm inactivity decreases protein synthesis and increases degradation of key contractile proteins, resulting in diaphragmatic force loss.3,15,17–21Protein degradation is mediated via oxidative stress,15,22,23apoptosis,24and proteasome proteolysis.5,25–27Few studies evaluating VIDD in vivo mainly with animals have been published3,15,21, although one was with humans.2Some other acute situations (i.e ., myorelaxants, sepsis, acute hypercapnic acidosis, and others) seem to induce VIDD in animal studies.14,28,29Sepsis may be associated with VIDD in the intensive care unit, although mainly explored through experimental studies. Several studies described the mechanisms leading to the VIDD, showing inflammation, proteolysis, or nitric oxide pathways.28,30–32Although one study demonstrated that CMV decreases VIDD compared with spontaneous breathing in septic rats, to the best of our knowledge,33none compared assisted versus CMV.
In this study, we confirmed that prolonged CMV decreased diaphragmatic contractile force in vivo in a healthy piglet model. In comparison with baseline, diaphragmatic force-generating capacity decreased by 25% after 72 h of ventilation in the CMV group (figs. 3A and 4).
Although it is widely accepted that VIDD occurs after several days of CMV, few countermeasures have been tested. Recently, it has been shown that administration of Trolox,34an antioxidant, glutamine,35or leupeptin,25which decreases calpain and cathepsin activity (proteases), prevents VIDD during mechanical ventilation in a rat model. Although medications to prevent or treat VIDD may be a point of interest in the future, ventilator strategies that allow diaphragmatic contractions seem to protect against VIDD.
Indeed, Sassoon et al .4reported the beneficial effect of assisted controlled ventilation on in vitro contractile properties of the diaphragm and the decrease in the muscle atrophy factor-box messenger RNA, a marker of gene atrophy in comparison with total CMV in rabbits. Gayan-Ramirez et al .1also reported the beneficial effect of intermittent spontaneous breathing on in vitro contractility and myogenesis regulatory factors in the diaphragm. Recently, we reported that PSV was effective in reducing diaphragm proteolysis and inhibition of protein synthesis compared with continuous mechanical ventilation.5This study is the first one performed in an in vivo model of piglet, a large animal whose physiology is similar to human beings, to report that maintaining spontaneous activity of the diaphragm with ASV may protect diaphragm against VIDD evaluated both in vivo (contractility) and in vitro (atrophy and fibers repartition). With the ASV mode, the ventilator software uses a modified version of the equation derived by Otis et al .7to minimize the work of breathing.8,12,13,36,37For subjects who are unable to trigger a breath, the ventilator generates a pressure-controlled cycle, automatically adjusting inspiratory pressure and time to achieve a target VTand a target RR. When the spontaneous activity of the diaphragm is sufficient to trigger a breath, the ventilator generates pressure-support breaths and, if necessary, delivers additional pressure-controlled breaths. In the ASV mode, similar to PSV, when an apnea occurs, the ventilator switches to the CMV mode but with an option that allows the patient to return to an assisted mode and triggers breaths.
In this 72 consecutive-hour study, the ASV mode allowed for a high proportion of pressure-supported cycles but ensured initiation of the apnea-induced CMV mode when necessary, with an optional return to PSV. In the ASV group, piglets were ventilated on pressure support more than 80% of the time, although they were ventilated in CMV mode the remainder of the time. On the contrary, in the CMV group, we could identify diaphragmatic activity during lesser than 5% of all breaths on the trend graphs of the Galileo® ventilator.
In this study, the piglets ventilated in ASV mode did not show any decrease in Pdi during the study period, contrary to those ventilated with CMV mode in whom Pdi decreased by 25% after 60 h of ventilation. We speculate that this protective effect is related to the spontaneous activity of the diaphragm because we did not compare ASV with pressure support or any other spontaneous mode.
Some study limitations must be pointed out. First, we did not compare our results with a control group without anesthesia, or mechanical ventilation, because large animals must be anesthetized for procedures such as phrenic stimulation and Pdi recording. Second, we did not compare ASV with PSV mode and therefore cannot conclude on the relative effects of spontaneous ventilation versus the specificity of the ASV mode. However, ASV may have beneficial effects compared with PSV because of the theoretical decrease in the work of breathing with the equation software derived by Otis et al . and the possibility of an automated backup from the apnea ventilation to ASV (i.e ., spontaneous ventilation mode) and vice versa , which allowed piglets to breathe spontaneously as much as possible. Third, although piglet respiratory muscles are close to those of humans, this study was limited because we compared the effect of CMV with that of ASV on healthy diaphragm muscles. Nevertheless, we can speculate that the reported alteration of Pdi would be even worse in pathologic situations, such as sepsis, that induce VIDD per se .28,38,39In the CMV group, sedation level was more important than that in the ASV group to neutralize the centers of breathing. Although we cannot eliminate a direct effect of sedation on Pdi, a recent review on VIDD stated that the effect of sedation on diaphragm function was clearly lower than the direct effect of mechanical ventilation.40Furthermore, in the clinical situation, sedation is frequently necessary to ensure patient–ventilator synchrony with the CMV mode and may lengthen the ventilation weaning process, thereby promoting VIDD.41,42Although this point represents clearly a methodological limit of our study, it reflects the clinical interaction among sedation, mechanical ventilation, and weaning.42,43
In conclusion, this study reports that maintaining diaphragmatic contractile activity during 72 consecutive hours with the ASV mode may protect the diaphragm against the deleterious effect of total CMV in healthy piglets. ASV prevented the in vivo alteration of diaphragmatic contractility and in vitro diaphragmatic atrophy, contrary to total CMV, which was associated with both diaphragmatic atrophy and in vivo alteration of the diaphragm contractility. Further studies, including physiologic human studies, are required to more fully understand the effect of ASV on diaphragmatic function.