- Email: [email protected]
Sedation with midazolam worsens the diaphragm function than dexmedetomidine and propofol during mechanical ventilation in rats
Biomedicine & Pharmacotherapy 121 (2020) 109405
Contents lists available at ScienceDirect
Biomedicine & Pharmacotherapy journal homepage: www.elsevier.com/locate/biopha
Original article
Sedation with midazolam worsens the diaphragm function than dexmedetomidine and propofol during mechanical ventilation in rats
T
Shao-Ping Li1, Xian-Long Zhou1, Yan Zhao* 169 Donghu Road, Emergency Center, Zhongnan Hospital of Wuhan University, Wuhan, Hubei, 430071, China
ARTICLE INFO
ABSTRACT
Keywords: Mechanical ventilation Diaphragm dysfunction Midazolam Dexmedetomidine Propofol
Background: Mechanical ventilation (MV) is identified as an independent contributor to diaphragmatic atrophy and contractile dysfunction. Appropriate sedation is also essential during MV, and anesthetics may have direct adverse effects on the diaphragm. However, there is a lack of research into the effects of different anesthetics on diaphragm function during MV. Objectives: In the present study, we aim to examine the effect of midazolam, dexmedetomidine, and propofol on diaphragm function during MV. Design: Animal study. Setting: University research laboratory. Subjects: Male Wistar rats. Interventions: Animals were experienced 12 h of MV or spontaneous breathing (SB) with continuous anesthetics infusion. Diaphragm contractile properties, cross-sectional areas, microcirculation, oxidative stress, and proteolysis were examined. Measurements and main results: Diaphragmatic specific force was markedly reduced in the midazolam group compared with the dexmedetomidine (−60.4 ± 3.01%, p < 0.001) and propofol group (−58.3 ± 2.60%, p < 0.001) after MV. MV sedated with midazolam induced more atrophy of type II fibers compared with dexmedetomidine (−21.8 ± 2.11%, p = 0.0001) and propofol (−8.2 ± 1.53%, p = 0.003). No significant differences of these indices were found in the midazolam, dexmedetomidine, and propofol groups under SB condition (all p > 0.05, respectively). Twelve hours of MV resulted in a time dependent reduction in diaphragmatic functional capillary density (PB −25.1%, p = 0.0001; MZ −21.6%, p = 0.0003; DD −15.2%, p = 0.022; PP −24.8%, p = 0.0001, respectively), which did not occur in the gastrocnemius muscle. The diaphragmatic lipid peroxidation adducts 4-HNE and HIF-1α levels were significantly lower in dexmedetomidine group and propofol group compared to midazolam group (p < 0.05, respectively). Meanwhile, the catalase and SOD levels were also relatively lower (p < 0.05, respectively) in midazolam group compared to dexmedetomidine group and propofol group. Conclusions: Twelve hours of mechanical ventilation during midazolam sedation led to a more severe diaphragm dysfunction than dexmedetomidine and propofol, possibly caused by its relative weaker antioxidant capacity.
1. Introduction Mechanical ventilation (MV) is widely used on patients needing lifesupporting conditions. However, approximately 80% of patients requiring prolonged MV (≥12 h) exhibit a profound and rapid reduction
in diaphragmatic force-generating capacity [1]. Ventilator-induced diaphragmatic dysfunction (VIDD) is a time-dependent decrease of diaphragm strength and/or muscle atrophy after initiation of MV. It is still hard to make diagnose although advanced techniques (ultrasound, pressure, and electromyography) applied [2]. High risk of
Abbreviations: MV, mechanical ventilation; SB, spontaneous breathing; PB, pentobarbital; MZ, midazolam; DD, dexmedetomidine; PP, propofol; ROS, reactive oxidative species; VIDD, ventilator-induced diaphragmatic dysfunction; PIP, peak inspiration pressure; MAP, mean arterial pressure; PaO2, arterial partial pressure of O2; PaCO2, arterial partial pressure of CO2; TVD, total vascular density; MFI, microcirculatory flow index; HI, heterogeneity index; FCD, functional capillary density; VILI, ventilation induced lung injury; RVD, rat-ventilator dyssynchrony; SDF, sidestream dark field ⁎ Corresponding author. E-mail address: [email protected] (Y. Zhao). 1 These authors contributed equally to this research. https://doi.org/10.1016/j.biopha.2019.109405 Received 24 July 2019; Received in revised form 25 August 2019; Accepted 28 August 2019 0753-3322/ © 2019 The Authors. Published by Elsevier Masson SAS. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
complications such as a lower daily probability of liberation from ventilation and prolonged sedatives consumption make it worth getting more investigations [3]. Studies in animals and humans revealed that multiple biochemical alterations such as increased production of reactive oxygen species (ROS), decreased protein synthesis, enhanced proteolysis, and apoptosis existed in the injured diaphragm [4–6]. A growing body of evidences suggest that increased diaphragmatic ROS production is essential for the trigger of molecular cascades [7]. Indeed, VIDD could be attenuated by ROS inhibitor or scavenger was already demonstrated by other authors [8] and our previous animals study [9]. Anesthetics such as propofol, dexmedetomidine, and midazolam were characterized with antioxidant function in both animals and humans [10–12]. It seems that using these sedatives could protect the diaphragm from oxidative injury. However, these anesthetics may have different effects on the diaphragm. In animals, propofol exhibited its peroxidation inhibition role in preserving the impaired diaphragm contractile profiles induced by septic peritonitis [11]. Contrarily, propofol itself as an independent factor contributed to the loss of the diaphragmatic strength and myofibers atrophy in spontaneously breathed condition [13]. Moreover, propofol decreased diaphragmatic electrical activity during the neutrally adjusted ventilatory assisted (NAVA) ventilation was also reported in humans [14]. The central α2agonist dexmedetomidine reduced oxidative stress and caspase-3 expression caused by ischemia-reperfusion injury in skeletal muscle [15], and attenuated muscle wasting associated with sepsis through a reduction in proteolysis [16]. However, dexmedetomidine impaired diaphragmatic function and increased diaphragmatic oxidative stress (200% dexmedetomidine vs. 73% pentobarbital) after MV [17]. The certain antioxidant capacities of propofol and dexmedetomidine did not against, even increased the oxidative stress in the diaphragm derived from MV. Midazolam could prevent cells death by inhibiting the cleaved Caspase-3 activity and oxidative stress injury [12]. But midazolam enhanced the dTc-induced diaphragmatic twitch depression in sepsis rats [18], and potentially reduced diaphragmatic activity in humans were reported [19]. Different from other forms of acquired diaphragmatic dysfunction, VIDD is potentially be treated by simple measures such as sedative regimens selection and ventilator settings [2]. In clinical practice, appropriate anesthetics is essential for maintaining adequate sedation during prolonged MV. A clear understanding of the effects of different anesthetics on the diaphragm is important for the VIDD medical treating strategy. So, a study to compare the effect of midazolam, propofol, and dexmedetomidine on diaphragm function during MV is needed.
Heavy Chain (MY-32), anti-Slow Skeletal Myosin Heavy Chain (NOQ7.5.4D), anti-Laminin were purchased from Abcam (Shanghai, China). SlowFadeTM Gold Antifade Mountant with DAPI and secondary antibodies including Alexa Fluor 546 conjugated donkey anti-mouse, Alexa Fluor 488 donkey anti-goat were purchased from Invitrogen (Shanghai, China). 2.3. Experimental design Two parallel animal experiments were performed in the present study: Experiment 1 (SB group): Animals breathed spontaneously with continuous infusion of pentobarbital (PB) (n = 6), midazolam (MZ) (n = 6), dexmedetomidine (DD) (n = 6) or propofol (PP) (n = 6) for 12 h. Experiment 2 (MV group): Animals were mechanically ventilated with continuous infusion of pentobarbital (PB) (n = 5), midazolam (MZ) (n = 5), dexmedetomidine (DD) (n = 5) or propofol (PP) (n = 5) for 12 h. It has been demonstrated that pentobarbital and the carrier of propofol (Lipovenous 10%) did not affect the diaphragm force [13,17], so we set the pentobarbital sedated groups as the sham. And no more animals were performed to examine the effect of the lipovenous in our study. Besides, the traditional sedation depth evaluation method (foot reflex, corneal reflex, and arterial blood pressure) was used in SB group; a new method (negative foot/corneal reflex, normal arterial blood pressure, peak inspiration pressure (PIP) ≤ 8 cm H2O) was used in MV group. Detail reasons about developing the new sedation depth evaluation method were stated in the appendix. 2.4. Methods All animals underwent a tracheotomy, right carotid artery, and jugular vein cannulation after peritoneal injection of sodium pentobarbital (60 mg•kg−1). Heart rate and mean arterial pressure (MAP) were continuously measured using biologic data acquisition and analysis system (BL-420 F, Wuhan, China). Arterial blood samples were taken every 4 h for blood cell counts (Pentra MS CRP, HORIBA, Japan) and blood gas (i-STAT300 G, Abbott, USA) examination. In addition, venous blood glucose was measured (ONE TOUCH, Johnson, USA) every 4 h at the end of the tail. Body fluid homeostasis was ensured by continuous administrating electrolyte solution (Baxter, Deerfield, IL) through the jugular vein (1 ml•kg−1 h−1). The body temperature was maintained at 37 °C using a temperature controller (BP-2010A, Softron, Japan) in a supine position. Bladder expression, eyes lubrication, and airway mucus removal were operated periodically. The ratio of the airoxygen gas mixture was adjusted to maintain the arterial partial pressures of oxygen (PaO2) between 80 mmHg and 100 mmHg. Animals in the MV group were ventilated using a volume controlled small-animal ventilator (VentElite, Harvard Apparatus, USA). Anesthetics dosages were administered to maintain a relatively steady sedation depth. The respiration waves and PIP were recorded by a connected computer. The tidal volume was set at 6–7 ml•kg−1 body weight with an inspiration/expiration ratio at 1:1. Respiratory rate was 60–70 breaths/minute and adjusted to maintain arterial partial pressures of carbon dioxide (PaCO2) at less than 50 mmHg. No positive endexpiratory pressure was used. The gas mixture was humidified and maintained at 37 °C.
2. Materials and methods 2.1. Animals Specific pathogen-free male Wistar rats weighing about 300 g were purchased from WeiTongLiHua experimental animals Co. Ltd (Beijing, China), and feed in the ABSL-III laboratory of Wuhan University. The study was approved by the Animal Experiment Center of Zhongnan Hospital of Wuhan University (Wuhan, China) with the serial number: 2018039. All procedures were carried out in compliance with the Institutional Animal Care and Use Committee. 2.2. Anesthetics and reagents
2.4.1. Microcirculatory measurements A Sidestream Dark Field (SDF) imaging device (MicroSee, Guangzhou, China) was used to measure the diaphragm microcirculation. The measurement points were set at the time of 30 min (T0), 6 h (T6) and 12 h (T12) after MV initiation. A skin incision 1 cm long was performed via a midline laparotomy to expose the cavity without obvious bleeding. The probe was inserted into the cavity and placed on the surface of the costal and crural diaphragm segments. The
Sodium pentobarbital (Saiyisi, Wuhan, China), dexmedetomidine (Aibeining, Jiangsu, China), propofol 2% (Diprifusor, AstraZeneca, Italy), and midazolam (Liyuexi, Jiangsu, China) were prepared for continuous infusion. Krebs-Henseleit bicarbonate buffer was purchased from M&C Gene Technology Ltd. (Beijing, China). Primary antibodies including anti-atrogin-1, anti-4-HNE, anti-catalase, anti-SOD1/2, antiAkt, anti-p-Akt, anti-Murf-1, anti-HIF-1α, anti-Fast Myosin Skeletal 2
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
manipulation was gently applied without pressure on the diaphragm and liver. Five sequences of 20 s each from different adjacent areas were recorded and stored under an alpha-numeric code. All sequences were acquired by the same manipulator and analyzed by a second blinded investigator. The total vascular density (TVD), microcirculatory flow index (MFI), flow heterogeneity index (HI), and functional capillary density (FCD) were analyzed according to the international recommended guidelines [20]. In addition, the right hindlimb gastrocnemius muscle microcirculation was also measured. Twelve hours later, a strip of costal diaphragm muscle was dissected for in vitro contractile measurements. The remaining diaphragm was stored to further determine biochemical and pathological examination. The lung was also removed and fixed for ventilation induced lung injury (VILI) evaluation.
then separated via SDS-PAGE through a stacking (5%) and separation (8%) gel. Subsequently, proteins were transferred to polyvinylidene fluoride (PVDF) membranes (200 mA for 1 h). The membranes were incubated with primary antibodies at 4 °C overnight. After washing with TBST, the membranes were probed with HRP-labeled secondary antibodies. Then the membranes were detected via chemiluminescence and image densitometry analysis with a luminescent image analysis system (Tanon-6200, China). 2.5. Statistical analysis Data are shown as number or mean ± standard (SD). Normal distribution was assessed with the Shapiro-Wilk test. The primary outcome was the in vitro total tension of the diaphragm and gastrocnemius. The 2-tailed Student’s t test was used for single biochemical or physiological observation comparison. Post-hoc analysis was performed as described if statistical significance (p ≤ 0.05) was achieved. Multiple responses were analyzed by one-way or two-way ANOVA. The relationship between MAP and muscle microcirculation was investigated using linear regression and Spearman correlation. Calculations were performed using Graphpad Prism 6.0.
2.4.2. Diaphragm contractility The method was elaborated in our prior study [9]. A system (ALCMPA2000, Shanghai, China) including organ bath chambers, transducer, and analysis software was used. The Krebs-Henseleit (K-H) solution was bubbled with a gas mixture of 95% O2-5% CO2 and maintained at 37 °C and pH 7.4. Optimal muscle length (Lo) for peak twitch force was established and the following parameters were performed at Lo: 1) maximum twitch force; 2) maximal tetanic force; 3) force-frequency relationship. The weight mass, length, width, and thickness were measured at Lo. All forces were normalized for CSA.
3. Results 3.1. General characteristics of animals in all groups
2.4.3. Immunofluorescence (IF) staining Pieces embedded in Optimal Cutting Temperature (OCT) were cut into 8-μm slices under −20 °C using a frozen slicer (CM1900, Leica, Germany). Two adjacent sections were stained with either mouse antislow myosin heavy chain antibody or anti-fast myosin heavy chain antibody. Additionally, rabbit anti-laminin antibody was included in each of these stains. After reaction with a mixture of Alexa Fluor 546 conjugated donkey anti-mouse and Alexa Fluor 488 donkey anti-goat secondary antibody, the sections were sealed with SlowFadeTM Gold Antifade Mountant with DAPI. Images were obtained using an OLYMPUS IX73 microscope (Olympus Co., Japan) and CSA was calculated with ImageJ software (Fiji) with at least 200 fibers per animal.
At the end of the experiments, MAP was significantly lower in comparison to baseline in all groups. Data of MAP, heart rate, blood gas, lactate, and glucose before euthanization were not significantly differences in all the MV subgroups (p > 0.05, respectively, Table 1) and the SB subgroups (p > 0.05, respectively, Table S1). In addition, differences in the blood cell counts were also insignificant in subgroups of MV and SB. Totally, no obvious glucose disorder, acid-base imbalance and systemic infections in animals in the present study. Anesthetics consumption dosages were significantly lower in the SB group compared to the MV group (p < 0.05, respectively). Table 1 Characteristics of the arterial blood gas, vital signs, blood cell counts, and total sedatives dosages used at the end of experiments in MV group.
2.4.4. Hematoxylin and eosin (HE) staining Tissues fixed with 4% paraformaldehyde were chopped into 0.5 cm × 0.5 cm × 1 cm pieces. Samples underwent dehydration, wax immersion and were then embedded in paraffin (ASP200S, Leica, Germany) for cutting into slices (RM2245, Leica, Germany). 5 μm sections were routinely stained with HE (Baso, Wuhan, China). Images were obtained using an OLYMPUS IX73 microscope (Olympus Co., Japan).
Group Arterial blood gas pH PaCO2 (mmHg) PaO2 (mmHg) Lactate (mmol/L) HCO3− (mmol/L) BEecf (mmol/L) Vital signs Heart rate (beats/ min) Rectal temperature (°C) MAP (mmHg) Glucose (mmol/ L) Blood cell counts Erythrocytes (x1012/L) Hemoglobin (g/ L) Neutrophils (%) Lymphocytes (%) Sedatives dosages (mg)
2.4.5. Transmission electron microscope (TEM) Tissues fixed with 2.5% glutaraldehyde (Servicebio, China) were chopped into 0.1 cm2 × 1 cm pieces. Then the tissues were rinsed, dehydrated, embedded, sliced, stained, and subsequently investigated under a transmission electron microscope (Tecnai G2 20 TWIN, FEI, USA). 2.4.6. Western blotting The protein level of 4-hydroxynonenal (4-HNE), hypoxia inducible factor-1α (HIF-1α), catalase, superoxide dismutase 1/2 (SOD1/2), protein kinase B (Akt), muscle ring finger-1 (Murf-1), atrogin-1, LC3BII/I were measured by SDS-PAGE and immunoblotting as previously described [9]. Briefly, 30 mg of diaphragm samples were homogenized in 200 μl RIPA lysis buffer with cocktail tablets (Servicebio, China) and centrifuged at 1200 RCF units for 15 min at 4 °C. The supernatant was then collected and its protein concentration was determined using the Total Protein Assay kit (BCA method, Beyotime Biotechnology, China). Eighty micrograms of protein from each of the diaphragm samples were
PB (n = 5)
MZ (n = 5)
DD (n = 5)
PP (n = 5)
7.42 ± 0.07 34 ± 4 104 ± 11 1.3 ± 0.2 25.6 ± 2.8 3±1
7.44 ± 0.05 33 ± 3 106 ± 8 1.5 ± 0.3 26.6 ± 2.1 3±1
7.43 ± 0.04 33 ± 4 110 ± 10 1.4 ± 0.3 24.6 ± 2.5 4±1
7.43 ± 0.04 35 ± 4 107 ± 7 1.1 ± 0.2 25.8 ± 2.6 3±2
386 ± 19
390 ± 12
394 ± 14
382 ± 14
37.3 ± 0.2
37.1 ± 0.1
37.0 ± 0.1
37.3 ± 0.3
103 ± 5 5.2 ± 0.8
105 ± 4 5.0 ± 0.7
107 ± 4 5.6 ± 0.9
104 ± 5 5.4 ± 0.4
7.1 ± 0.3
7.5 ± 0.8
6.9 ± 0.4
7.0 ± 0.3
161 ± 8
158 ± 9
165 ± 6
160 ± 6
46.1 ± 3.1 36.9 ± 3.7 47 ± 3
44.8 ± 2.6 35.3 ± 4.0 18 ± 3
44.1 ± 1.7 37.4 ± 3.9 0.106 ± 0.005
45.2 ± 2.3 36.7 ± 4.2 165 ± 3
Data are expressed as number or mean ± SD; PB = pentobarbital group; DD = dexmedetomidine group; PP = propofol group; MZ = midazolam group. BEecf = base excess of extracellular fluid; MAP = mean arterial pressure. 3
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
Fig. 1. Force frequency curves and cross sectional areas of diaphragm in SB and MV groups (mean ± SD). Diaphragm force-frequency curve in SB group (a) and MV group (b); cross sectional areas (CSA) of diaphragm in SB group (c) and MV group (d). Groups represent: PB = pentobarbital group; DD = dexmedetomidine group; PP = propofol group; MZ = midazolam group; white box = type I fibers; grey box = type II fibers. * p < 0.05 vs. pentobarbital group, # p < 0.05 vs. midazolam group.
3.2. Diaphragm contractile properties and cross-sectional areas
3.3. Diaphragm microcirculation in the MV group
Twelve hours of MV or SB resulted in a significant reduction in diaphragm specific force in MZ, DD, and PP groups compared to PB. The discrimination between MZ, DD, and PP were not significant (p > 0.05, respectively) under the SB condition (Fig. 1a). However, midazolam more clearly induced force reduction compared to propofol (−58.5 ± 10.2%, p= 0.0001) and dexmedetomidine (−64.0 ± 10.5%, p = 0.0001) after MV at stimulation frequency ranging from 10 to 120 Hz (Fig. 1b). Although the diaphragmatic forces were impaired, midazolam, dexmedetomidine, and propofol did not induce further diaphragm atrophy compared to pentobarbital under SB condition (Figs. 1c, S1). Dexmedetomidine and propofol were more effective in protecting the fibers from atrophy after MV compared to midazolam, especially in the Type II fibers (p = 0.0001, respectively) (Figs. 1d and 2 ). Different from diaphragm, fiber size and contraction force in gastrocnemius were kept unchanged in MZ, DD, and PP groups compared to PB after12 h of MV or SB (Fig. S2-S4).
No statistical differences of PIP and MAP existed between all the MV groups during the sequences manipulation (Table 2). During the whole process of the measurements (T0, T6, T12), red cells moved fluently in the vessels (diameter ≤ 20 μm), and no obvious of heterogenetic blood flow and vessel occlusion (10 μm ≤ diameter ≤ 20 μm) were observed despite the diaphragm disused (Video S1). There were no significant differences of TVD, MFI, and HI between all the groups at the time of T0, T6, and T12. But a time-dependent decline in FCD was observed in all the groups (Fig. 3a–c). FCD decreased significantly (PB −25.1%, p = 0.0001, MZ −21.6%, p = 0.0003, DD −15.2%, p = 0.022, PP −24.8%, p = 0.0001, respectively) from T0 to T12 (Fig. 3d). Dexmedetomidine was more efficient in preserving the perfused capillary than midazolam or propofol, especially at T12 (p = 0.006, relative to propofol). Weak relationship between FCD and MAP was observed from T0 to T6 (R2 = 0.263, p < 0.001) but not T6 to T12 (R2 = 0.096, p = 0.016) (Fig. S5). Unlike the diaphragm, alterations of microcirculation of the diaphragm were not observed in the gastrocnemius muscle.
Fig. 2. Representative immunofluorescent staining of type I (above) and type II (below) fibers in diaphragm in MV group. Groups represent: PB = pentobarbital group; DD = dexmedetomidine group; PP = propofol group; MZ = midazolam group. 4
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
using midazolam (4.5 ± 0.6%, p = 0.0001) or dexmedetomidine (4.7 ± 0.7%, p = 0.0001) induced higher diaphragmatic autophagic activity than propofol (1.9 ± 0.3%), and these alterations were also presented in the autophagosome formation (Fig. 6).
Table 2 Characteristics of the peak inspiration pressure, mean arterial blood pressure at T0, T6 and T12. Group PIP (cmH2O) T0 T6 T12 MAP (mmHg) T0 T6 T12
PB (n = 5)
MZ (n = 5)
DD (n = 5)
PP (n = 5)
5.4 ± 1.6 5.2 ± 1.3 5.6 ± 1.8
5.8 ± 1.8 5.5 ± 1.4 5.3 ± 1.2
6.1 ± 2.0 5.1 ± 1.5 5.5 ± 1.4
5.4 ± 1.2 5.4 ± 1.6 5.6 ± 1.1
132.4 ± 5 105 ± 4 * 103 ± 5 *
130.1 ± 6 107 ± 5 * 105 ± 4 *
137.2 ± 5 101 ± 4 * 107 ± 4 *
134.8 ± 4 105 ± 4 * 104 ± 4 *
4. Discussion Our experiments provide important information regarding the role that anesthetics play in VIDD. This research systematically compares the effects of dexmedetomidine, propofol, and midazolam on diaphragm function under MV and SB in rats. Conclusively, midazolam, dexmedetomidine, and propofol all have direct adverse effects on the diaphragm contraction force under SB. MV sedated with midazolam results in a more severe diaphragm dysfunction than dexmedetomidine or propofol, possibly caused by its relative weak capacity in inhibiting the diaphragmatic ROS production or protein degradation. Propofol induced similar diaphragm dysfunction compared to dexmedetomidine. MV results in a time-dependent decrease of FCD in the disused diaphragm, and dexmedetomidine is more effective in preserving perfused capillary density than midazolam or propofol.
Data are expressed as number or mean ± SD. PB = pentobarbital group; DD = dexmedetomidine group; PP = propofol group; MZ = midazolam group. PIP = peak inspiration pressure; T0, T6, T12 = 30 min, 6 h, 12 h after mechanical ventilation initiation. * p < 0.05 vs. T0.
3.4. Oxidative stress in the MV group The lipid peroxidation adducts 4-HNE and HIF-1α levels were significantly lower in DD and PP compared to MZ (Fig. 4a-d) (p < 0.05, respectively). Meanwhile, the catalase (Fig. 4e) and SOD levels (Fig. 4f) were also relative lower (p < 0.05, respectively) in MZ compared to DD or PP. Midazolam did not inhibit or eliminate the ROS production in the diaphragm efficiently than dexmedetomidine and propofol.
4.1. Midazolam causes a more severe diaphragm dysfunction and oxidative stress than propofol or dexmedetomidine in ventilated animals As stated above, MV using midazolam sedation induced a more severe diaphragmatic atrophy and contractile dysfunction than dexmedetomidine and propofol. In prior experiments under pentobarbital anesthesia, oxidative stress was recognized as the main factor modifying up or down pathways involved in the process of VIDD [2]. We find dexmedetomidine and propofol but not midazolam significantly decreased the diaphragmatic ROS production compared to pentobarbital. Anesthetics protecting organisms by defending oxidative stress injury have already been reported in various models. Dexmedetomidine reduced oxidative stress caused by ischemia-reperfusion injury in skeletal muscle [14]. Propofol enhanced heme oxygenase expression and generated antioxidant activities because its molecular structure is
3.5. Protein degradation in MV group Regulator of the autophagy-lysosome pathway and the ubiquitin proteasome system-pAkt/Akt ratio was only significantly decreased in DD (−48.1%, p = 0.0001) compared to PB (Fig. 5a, b). Meanwhile, markers that acted as the indicators of ubiquitin proteasome system Murf-1 (Fig. 5c) and atrogin-1 (Fig. 5d) were decreased in DD and PP but not MZ compared to PB. Autophagic activity, represented by the ratio of LC3B-II/LC3B-I, were statistically increased in all the three groups compared to PB (p < 0.001, respectively) (Fig. 5e, f). Sedation
Fig. 3. Representative SDF images and summary of functional capillary density of rat diaphragm microcirculation. Images sequenced at 30 min (a), 6 h (b), and 12 h (c) after MV initiation in pentobarbital group (objective 5x, on screen 325x). Summary of functional capillary density of all MV groups (d). Values are displayed as mean ± SD. FCD = functional capillary density; T0, T6, T12 = 30 min, 6 h, 12 h after MV initiation. Groups represent: PB = pentobarbital group; DD = dexmedetomidine group; PP = propofol group; MZ = midazolam group. * p < 0.05 vs. T0.
5
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
Fig. 4. Assessment of oxidant and anti-oxidant indicators in the diaphragm of the MV group. Representative immunoblots of 4-hydroxynonenal (4-HNE) (a). Ratio of 4-HNE (b) (mean ± SD). Representative immunoblots of hypoxia induced factor-1α (HIF-1α), catalase, and superoxide dismutase (SOD1/2) (c). Ratio of HIF-1α (d), catalase (e), and SOD1/2 (f) (mean ± SD). Groups represent: PB = pentobarbital group; MZ = midazolam group; DD = dexmedetomidine group; PP = propofol group; GADPH = glycerinaldehyd-3-phosphat-dehydrogenase. * p < 0.05 vs. pentobarbital group, # p < 0.05 vs. midazolam group.
similar to vitamin E [21]. Midazolam prevented cells death from oxidative stress mediated by JNK-ERK pathway [12]. From our data, it is concluded that dexmedetomidine and propofol are superior to midazolam in inhibiting lipid peroxidation and increasing antioxidant enzymes activity in the disused diaphragm. And this finding is also supported by the fact that propofol and dexmedetomidine exhibit a superior antioxidant function to midazolam in humans [22]. Midazolam does not, at least not effectively exhibit its antioxidant property in ventilated diaphragm when compared to pentobarbital at sedation dosage. Our results demonstrate that the effects of anesthetics on diaphragm function during MV are associated with their antioxidant properties. Ubiquitin proteasome system activated by the increased oxidative stress was recognized as the main reason causing diaphragm fiber atrophy [23]. Consistent to the ROS level, Murf-1 and atrogin-1 expression levels were decreased differently in DD and PP but not MZ compared to PB. But the ratios of LC3B-II/LC3B-I are statistically increased in all the three groups compared to PB, especially in MZ and DD. Dexmedetomidine increased autophagy activity in VIDD was already been demonstrated [17]. Midazolam was also implied that could induce autophagy to control cell apoptosis [24]. Propofol is most efficient in inhibiting the diaphragmatic autophagic activity and this is in line with other studies [25]. In fact, the function of autophagy in the process of VIDD is still in controversial. It was reported that autophagy was necessary for VIDD, and increased autophagy stimulated by oxidative stress will further resulting in the increase of autophagy and ROS [26]. However, the other study indicated that autophagy induced by MV might instead be a beneficial adaptive response and was not responsible for diaphragm atrophy [27]. Our experiments provide evidence that autophagy is indeed involved in the development of VIDD, and anesthetics have different influences on the autophagy. But
whether these influences are beneficial or adverse for the diaphragm during VIDD is still unclear. 4.2. Dexmedetomidine induced diaphragm dysfunction but prevent muscle atrophy in a low level of oxidative stress and proteolytic activity under MV Dexmedetomidine was already shown to reequilibrate the oxidantantioxidant balance especially by enhancing antioxidant defense enzymes and inhibiting lipid peroxidation in various models [14,28,29]. Our data corroborate these studies. Currently, dexmedetomidine protects the diaphragm from MV is associated with its antioxidant function. Thomas et al concluded that dexmedetomidine led to a worsening of VIDD but diaphragmatic fiber atrophy was prevented even in the circumstance of increased oxidative stress and activated proteolytic pathways [17]. Contrarily, the level of 4-HNE, catalase, and SOD1 in the DD were significant lower compared with PB in our experiment. Thomas et al used a traditional sedation depth evaluation method in their experiments, which led to the prevalence of RVD during MV. We also demonstrated that the diaphragmatic oxidative stress level was significant higher in RVD positive rats compared to the negative ones after 12 h of MV under pentobarbital sedation. Their study could not exclude the extra oxidative stress produced by RVD when ventilation was carried out using traditional sedation depth method. Noticeably, the duration of MV in our experiment is half that of their study. VIDD occurred in 12 h of MV was proved by others [26], and our previous experiment [9]. Whether the additional 12 h’ MV with RVD induce a severer diaphragm dysfunction or higher oxidative stress level by dexmedetomidine sedation is still unknown. Our results indicate that, at least in 12 h, the antioxidant function of dexmedetomidine is reserved in VIDD, and more effective than propofol or midazolam.
6
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
Fig. 5. Assessment of proteasome and autophagy indicators in the diaphragm of the MV group. Representative immunoblots of phosphorylated/total Akt, Murf-1, and Atrogin-1 (a). Ratio of p-Akt/Akt (b), Murf-1 (c), and Atrogin-1 (d) (mean ± SD). Representative immunoblots of of LC3B-II/LC3B-I (e). Ratio of LC3B-II/LC3B-I (f) (mean ± SD). Groups represent: PB = pentobarbital group; MZ = midazolam group; DD = dexmedetomidine group; PP = propofol group; GADPH = glycerinaldehyd-3-phosphat-dehydrogenase. * p < 0.05 vs. pentobarbital group, # p < 0.05 vs. midazolam group.
4.3. Propofol induced similar diaphragm dysfunction compared to dexmedetomidine
Because it was already recognized that pentobarbital did not affect the diaphragm force, the conclusion that propofol did not protect against ventilator-induced diaphragmatic atrophy and the oxidative injury was quite not convincing without a “control” group- MV + pentobarbital. The authors neglected the fact that MV was independent of sedatives contributed to diaphragmatic atrophy and contractile dysfunction. Second, the duration of SB or MV in our experiment is half that of the prior study. Third, the mean propofol infusion dosages in our experiments (27 mg•kg−1 h−1 in SB group, 46 mg•kg−1 h−1 in MV group) are higher than the study used (20 mg•kg−1 h−1). But none of the animals did show any signs of propofol infusion syndrome at our dosages. Moreover, it was reported that a high dose (50 mg•kg−1 h−1) propofol was more beneficial in protecting the diaphragm from sepsis [11] or healthy rats [30]. Fourth, they also used the traditional sedation depth evaluation method which RVD could not be excluded in the MV group. Future study about the effects of different dosages and duration of propofol used on diaphragm function is needed.
Our data show that the diaphragmatic forces and fibers dimensions in PP are similar to DD. Likewise to dexmedetomidine, propofol is also effective in maintaining the diaphragmatic oxidant-antioxidant balance during MV. Our experiment further corroborates the hypothesis that diaphragm dysfunction induced by dexmedetomidine was similar to propofol [17]. But in an experimental model of 24 h of SB, propofol induced diaphragm atrophy directly without obvious oxidative stress injury [12]. This study also concluded that propofol did not protect against ventilator-induced diaphragmatic atrophy and oxidative injury. Differences of the experimental designs and anesthetics dosages used may be helpful to explain the contrary results. First, the set of “control” group is different. We set the animals breathed spontaneously or ventilated mechanically with continuous pentobarbital infusion which is much different from the direct euthanized after anesthetization.
Fig. 6. Representative transmission electrode microscope in diaphragm. Arrow = autophagosome. Groups represent: PB = pentobarbital group; MZ = midazolam group; DD = dexmedetomidine group; PP = propofol group. 7
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
4.4. MV induced diaphragm blood flow reduction most occurred in capillary
a lower antioxidant activity, higher lipid peroxidation and protein ubiquitination level. Sedation with midazolam worsens diaphragm function than dexmedetomidine and propofol during MV. Propofol and dexmedetomidine induced similar diaphragm dysfunction in a low level of oxidative stress and proteolytic activity under MV. MV induced diaphragm blood flow reduction, and this most occurred at the capillary level.
Prior to our study, it was already revealed that 6 h of MV resulted in a 75% reduction in whole blood flow in the disused diaphragm [31]. Our previous study also demonstrated that 12 h of MV resulted in a clear diaphragm blood flow reduction using the laser Doppler technique [9]. Now, we find the reduction of diaphragm blood flow during MV mostly occurred at the capillary using the SDF technique. And this finding is supported by the rapid declined perfused capillary density in the disused diaphragm. The components of the microvasculature (arterioles, capillaries, venules) are able to adapt to the function changed in the disused skeletal muscle. Structure of larger blood vessels (e.g., arterioles) appears to be preserved, and the major adaptation to disuse occurred at the capillary bed level [32,33]. Our results are kept in line with these studies. It was recognized that the microvascular circulation of the tissue was primarily dependent on the perfusion pressure of the organ. Although MAP was decreased during MV, however, FCD continued decreased even MAP was kept stable from T6 to T12. In addition, no relations were observed between MAP and FCD in all groups from T6 to T12. Conclusively, MV induced diaphragmatic FCD reduction was independent to the decrease of MAP. Although the diaphragmatic functional capillary density declined during ventilation, it possibly not the key factor contributed to the ventilator-induced diaphragm dysfunction.
Funding This work was supported by the Emergency Diagnostic and Therapeutic Center of Central China (Wuhan, Hubei, China). Declaration of Competing Interest The authors declare that there are no conflicts of interest. Acknowledgments Thank Yu-Qian Zhao and Qi Li (Intensive Care Unit of Zhongnan Hospital, Wuhan, China) for their help for Microsee (Guangzhou, China) manipulation. Thank Li Yang, Xiao-Hua Leng (Medical Research Center of Zhongnan Hospital, Wuhan, China) for their technique assistance. Thank Di Hu (MAZARS, Wuhan, China) for her help for data analysis.
5. Conclusions Compared to dexmedetomidine or propofol, midazolam resulted in A
Rat-Ventilator Dysyynchrony existed using a traditional sedation depth evaluation method In animals’ experimental models, a traditional method (foot reflex, corneal reflex and arterial blood pressure) was widely used for the sedation depths evaluation. For example, rats with negative foot/corneal reflex and normal arterial blood pressure were recognized as “adequate” sedated to receive prolonged mechanical ventilation (MV) [13,16]. However, during our pre-experiments of MV with four anesthetics sedation, we found ratventilator dyssynchrony (RVD) (evaluated by widely changed airway pressure, Video A1) existed through the whole experiments and could not be improved by ventilator settings adaption when rats’ sedation depth was evaluated by “adequate” level. We found RVD could be avoided using a new sedation depth evaluation method (negative foot/corneal reflex, normal arterial blood pressure, peak inspiration pressure (PIP) ≤ 8 cm H2O) with a continuous PIP and waveforms displayed ventilator (VentElite, Harvard Apparatus, USA). RVD results in a worsen diaphragmatic dysfunction and lung injury To examine the effect of RVD on diaphragmatic structural and contractile properties, we set a RVD group: rats (n = 5) experienced 12 h MV with continuous pentobarbital sedation using traditional sedation depth evaluation method. Subgroup PB in the MV group as the control (Con). Data of
Fig. A1. Representative diaphragmatic HE (left), diaphragmatic sirius red (middle) and lung HE (right) staining. Groups represent: Con = pentobarbital group; RVD = rat ventilator dyssynchrony group. 8
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
Fig. A2. Lipid peroxidation (4-hydroxynonenal, 4-HNE) in the diaphragm. Groups represent: Con = pentobarbital group; RVD = rat ventilator dyssynchrony group.
MAP, blood gas, lactate, and glucose before euthanization were not significantly different in the two groups (p > 0.05, respectively). There was also no obvious collagen rupture and inflammatory injury existed in both groups (Fig. A1). But RVD induced severe lung injury included more inflammatory cells and erythrocytes in alveolar walls/spaces, and more presence of ruptured and thickened alveolar walls than control. Meanwhile, RVD induced more clearly diaphragmatic contraction force reduction compared to Con (−41.4 ± 6.4%, p = 0.0001) at stimulation frequency ranging from 10 to 120 Hz. However, no significant differences of CSAs existed between the two groups (p = 0.401, respectively,). The 4-HNE level was also higher in RVD group compared to control (p = 0.003, respectively, Fig. A2). Appendix B. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.biopha.2019.109405.
[17] T. Breuer, C. Bleilevens, R. Rossaint, et al., Dexmedetomidine impairs diaphragm function and increases oxidative stress but does not aggravate diaphragmatic atrophy in mechanically ventilated rats, Anesthesiology 128 (4) (2018) 784–795. [18] E. Narimatsu, M. Aimono, Y. Nakayama, et al., The effects of midazolam and ketamine on D-tubocurarine-induced twitch depression in septic rat diaphragm, Res. Commun. Mol. Pathol. Pharmacol. 108 (5–6) (2000) 369–380. [19] H. Roze, A. Germain, V. Perrier, et al., Effect of flumazenil on diaphragm electrical activation during weaning from mechanical ventilation after acute respiratory distress syndrome, Br. J. Anaesth. 114 (2) (2015) 269–275. [20] C. Ince, E.C. Boerma, M. Cecconi, et al., Second consensus on the assessment of sublingual microcirculation in critically ill patients: results from a task force of the European Society of Intensive Care Medicine, Intensive Care Med. 44 (3) (2018) 281–299. [21] C. Liang, J. Cang, H. Wang, et al., Propofol attenuates cerebral ischemia/reperfusion injury partially using heme oxygenase-1, J. Neurosurg. Anesthesiol. 25 (3) (2013) 311–316. [22] C. Han, W. Ding, W. Jiang, et al., A comparison of the effects of midazolam, propofol and dexmedetomidine on the antioxidant system: a randomized trial, Exp. Ther. Med. 9 (6) (2015) 2293–2298. [23] S. Levine, T. Nguyen, N. Taylor, et al., Rapid disuse atrophy of diaphragm fibers in mechanically ventilated humans, N. Engl. J. Med. 358 (13) (2008) 1327–1335. [24] E.C. So, Y.C. Chen, S.C. Wang, et al., Midazolam regulated caspase pathway, endoplasmic reticulum stress, autophagy, and cell cycle to induce apoptosis in MA-10 mouse Leydig tumor cells, Onco. Ther. 9 (2016) 2519–2533. [25] H. Qiao, Y. Li, Z. Xu, et al., Propofol affects neurodegeneration and neurogenesis by regulation of autophagy via effects on intracellular calcium homeostasis, Anesthesiology 127 (3) (2017) 490–501. [26] A.J. Smuder, K.J. Sollanek, W.B. Nelson, et al., Crosstalk between autophagy and oxidative stress regulates proteolysis in the diaphragm during mechanical ventilation, Free Radic. Biol. Med. 115 (2018) 179–190. [27] I. Azuelos, B. Jung, M. Picard, et al., Relationship between autophagy and ventilator-induced diaphragmatic dysfunction, Anesthesiology 122 (6) (2015) 1349–1361. [28] H. Akpınar, M. Nazıroğlu, İ. Övey, et al., The neuroprotective action of dexmedetomidine on apoptosis, calcium entry and oxidative stress in cerebral ischemia-induced rats: contribution of TRPM2 and TRPV1 channels, Sci. Rep. 6 (2016) 37196. [29] V. Sen, A. Güzel, H.S. Şen, et al., Preventive effects of dexmedetomidine on the liver in a rat model of acid-induced acute lung injury, Biomed Res. Int. 2014 (2014) 621827. [30] S. Ulker, D. Tok, S. Kosay, et al., Effects of ketamine, thiopental sodium and propofol on muscle contractures in rat diaphragm in vitro, Int. J. Exp. Pathol. 72 (5) (1991) 527–532. [31] D.R.T. 3rd1, C.S. Bruells, J.N. Stabley, et al., Mechanical ventilation reduces rat diaphragm blood flow and impairs oxygen delivery and uptake, Crit. Care Med. 40 (10) (2012) 2858–2866. [32] A.B. Borisov, S.K. Huang, C. BM, Remodeling of the vascular bed and progressive loss of capillaries in denervated skeletal muscle, Anat. Rec. 258 (3) (2000) 292–304. [33] K. Tyml, M.-C. O, Structural and functional changes in the microvasculature of disused skeletal muscle, Front Biosci 6 (2001) D45–52.
References [1] A.C. Watson, P.D. Hughes, M. Louise Harris, et al., Measurement of twitch transdiaphragmatic, esophageal, and endotracheal tube pressure with bilateral anterolateral magnetic phrenic nerve stimulation in patients in the intensive care unit, Crit. Care Med. 29 (7) (2001) 1325–1331. [2] M. Dres, E.C. Goligher, L.M.A. Heunks, et al., critical illness-associated diaphragm weakness, Intensive Care Medicine 43 (10) (2017) 1441–1452. [3] E.C. Goligher, M. Dres, E. Fan, et al., Mechanical ventilation-induced diaphragm atrophy strongly impacts clinical outcomes, Am. J. Respir. Crit. Care Med. 197 (2) (2018) 204–213. [4] S.K. Powers, M.B. Hudson, W.B. Nelson, et al., Mitochondria-targeted antioxidants protect against mechanical ventilation-induced diaphragm weakness, Crit. Care Med. 39 (7) (2011) 1749–1759. [5] S.N. Hussain, M. Mofarrahi, I. Sigala, et al., Mechanical ventilation-induced diaphragm disuse in humans triggers autophagy, Am. J. Respir. Crit. Care Med. 182 (11) (2010) 1377–1386. [6] A.J. Smuder, M.B. Hudson, W.B. Nelson, et al., Nuclear factor-κB signaling contributes to mechanical ventilation-induced diaphragm weakness, Crit. Care Med. 40 (3) (2012) 927–934. [7] E. Zhu, S. CS, Ventilator-induced diaphragmatic vascular dysfunction, Crit. Care Med. 40 (10) (2012) 2914–2915. [8] J.M. McClung, D. Van Gammeren, M.A. Whidden, et al., Apocynin attenuates diaphragm oxidative stress and protease activation during prolonged mechanical ventilation, Crit. Care Med. 37 (4) (2009) 1373–1379. [9] X.L. Zhou, X.J. Wei, S.P. Li, et al., Interactions between Cytosolic Phospholipase A2 Activation and Mitochondrial Reactive Oxygen Species Production in the Development of Ventilator-Induced Diaphragm Dysfunction, Oxid. Med. Cell. Longev. 2019 (2019) 2561929. [10] J. Cui, H. Zhao, C. Wang, et al., Dexmedetomidine attenuates oxidative stress induced lung alveolar epithelial cell apoptosis in vitro, Oxid. Med. Cell. Longev. 2015 (2015) 358396. [11] K. Mikawa, K. Nishina, S. Kodama, Obara H, Propofol attenuates diaphragmatic dysfunction induced by septic peritonitis in hamsters, Anesthesiology 94 (4) (2001) 652–660. [12] G.Z. Li, H.L. Tao, C. Zhou, et al., Midazolam prevents motor neuronal death from oxidative stress attack mediated by JNK-ERK pathway, Hum. Cell 31 (1) (2018) 64–71. [13] C.S. Bruells, K. Maes, R. Rossaint, et al., Sedation using propofol induces similar diaphragm dysfunction and atrophy during spontaneous breathing and mechanical ventilation in rats, Anesthesiology 120 (3) (2014) 665–672. [14] A. Amigoni, G. Rizzi, A. Divisic, et al., Effects of propofol on diaphragmatic electrical activity in mechanically ventilated pediatric patients, Intensive Care Med. 41 (10) (2015) 1860–1861. [15] X. Dong, Q. Xing, Y. Li, et al., Dexmedetomidine protects against ischemia-reperfusion injury in rat skeletal muscle, J. Surg. Res. 186 (2014) 240–245. [16] M. Cheng, T. Gao, F. Xi, et al., Dexmedetomidine ameliorates muscle wasting and attenuates the alteration of hypothalamic neuropeptides and inflammation in endotoxemic rats, PLoS One 12 (3) (2017) e0174894.
9
Contents lists available at ScienceDirect
Biomedicine & Pharmacotherapy journal homepage: www.elsevier.com/locate/biopha
Original article
Sedation with midazolam worsens the diaphragm function than dexmedetomidine and propofol during mechanical ventilation in rats
T
Shao-Ping Li1, Xian-Long Zhou1, Yan Zhao* 169 Donghu Road, Emergency Center, Zhongnan Hospital of Wuhan University, Wuhan, Hubei, 430071, China
ARTICLE INFO
ABSTRACT
Keywords: Mechanical ventilation Diaphragm dysfunction Midazolam Dexmedetomidine Propofol
Background: Mechanical ventilation (MV) is identified as an independent contributor to diaphragmatic atrophy and contractile dysfunction. Appropriate sedation is also essential during MV, and anesthetics may have direct adverse effects on the diaphragm. However, there is a lack of research into the effects of different anesthetics on diaphragm function during MV. Objectives: In the present study, we aim to examine the effect of midazolam, dexmedetomidine, and propofol on diaphragm function during MV. Design: Animal study. Setting: University research laboratory. Subjects: Male Wistar rats. Interventions: Animals were experienced 12 h of MV or spontaneous breathing (SB) with continuous anesthetics infusion. Diaphragm contractile properties, cross-sectional areas, microcirculation, oxidative stress, and proteolysis were examined. Measurements and main results: Diaphragmatic specific force was markedly reduced in the midazolam group compared with the dexmedetomidine (−60.4 ± 3.01%, p < 0.001) and propofol group (−58.3 ± 2.60%, p < 0.001) after MV. MV sedated with midazolam induced more atrophy of type II fibers compared with dexmedetomidine (−21.8 ± 2.11%, p = 0.0001) and propofol (−8.2 ± 1.53%, p = 0.003). No significant differences of these indices were found in the midazolam, dexmedetomidine, and propofol groups under SB condition (all p > 0.05, respectively). Twelve hours of MV resulted in a time dependent reduction in diaphragmatic functional capillary density (PB −25.1%, p = 0.0001; MZ −21.6%, p = 0.0003; DD −15.2%, p = 0.022; PP −24.8%, p = 0.0001, respectively), which did not occur in the gastrocnemius muscle. The diaphragmatic lipid peroxidation adducts 4-HNE and HIF-1α levels were significantly lower in dexmedetomidine group and propofol group compared to midazolam group (p < 0.05, respectively). Meanwhile, the catalase and SOD levels were also relatively lower (p < 0.05, respectively) in midazolam group compared to dexmedetomidine group and propofol group. Conclusions: Twelve hours of mechanical ventilation during midazolam sedation led to a more severe diaphragm dysfunction than dexmedetomidine and propofol, possibly caused by its relative weaker antioxidant capacity.
1. Introduction Mechanical ventilation (MV) is widely used on patients needing lifesupporting conditions. However, approximately 80% of patients requiring prolonged MV (≥12 h) exhibit a profound and rapid reduction
in diaphragmatic force-generating capacity [1]. Ventilator-induced diaphragmatic dysfunction (VIDD) is a time-dependent decrease of diaphragm strength and/or muscle atrophy after initiation of MV. It is still hard to make diagnose although advanced techniques (ultrasound, pressure, and electromyography) applied [2]. High risk of
Abbreviations: MV, mechanical ventilation; SB, spontaneous breathing; PB, pentobarbital; MZ, midazolam; DD, dexmedetomidine; PP, propofol; ROS, reactive oxidative species; VIDD, ventilator-induced diaphragmatic dysfunction; PIP, peak inspiration pressure; MAP, mean arterial pressure; PaO2, arterial partial pressure of O2; PaCO2, arterial partial pressure of CO2; TVD, total vascular density; MFI, microcirculatory flow index; HI, heterogeneity index; FCD, functional capillary density; VILI, ventilation induced lung injury; RVD, rat-ventilator dyssynchrony; SDF, sidestream dark field ⁎ Corresponding author. E-mail address: [email protected] (Y. Zhao). 1 These authors contributed equally to this research. https://doi.org/10.1016/j.biopha.2019.109405 Received 24 July 2019; Received in revised form 25 August 2019; Accepted 28 August 2019 0753-3322/ © 2019 The Authors. Published by Elsevier Masson SAS. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
complications such as a lower daily probability of liberation from ventilation and prolonged sedatives consumption make it worth getting more investigations [3]. Studies in animals and humans revealed that multiple biochemical alterations such as increased production of reactive oxygen species (ROS), decreased protein synthesis, enhanced proteolysis, and apoptosis existed in the injured diaphragm [4–6]. A growing body of evidences suggest that increased diaphragmatic ROS production is essential for the trigger of molecular cascades [7]. Indeed, VIDD could be attenuated by ROS inhibitor or scavenger was already demonstrated by other authors [8] and our previous animals study [9]. Anesthetics such as propofol, dexmedetomidine, and midazolam were characterized with antioxidant function in both animals and humans [10–12]. It seems that using these sedatives could protect the diaphragm from oxidative injury. However, these anesthetics may have different effects on the diaphragm. In animals, propofol exhibited its peroxidation inhibition role in preserving the impaired diaphragm contractile profiles induced by septic peritonitis [11]. Contrarily, propofol itself as an independent factor contributed to the loss of the diaphragmatic strength and myofibers atrophy in spontaneously breathed condition [13]. Moreover, propofol decreased diaphragmatic electrical activity during the neutrally adjusted ventilatory assisted (NAVA) ventilation was also reported in humans [14]. The central α2agonist dexmedetomidine reduced oxidative stress and caspase-3 expression caused by ischemia-reperfusion injury in skeletal muscle [15], and attenuated muscle wasting associated with sepsis through a reduction in proteolysis [16]. However, dexmedetomidine impaired diaphragmatic function and increased diaphragmatic oxidative stress (200% dexmedetomidine vs. 73% pentobarbital) after MV [17]. The certain antioxidant capacities of propofol and dexmedetomidine did not against, even increased the oxidative stress in the diaphragm derived from MV. Midazolam could prevent cells death by inhibiting the cleaved Caspase-3 activity and oxidative stress injury [12]. But midazolam enhanced the dTc-induced diaphragmatic twitch depression in sepsis rats [18], and potentially reduced diaphragmatic activity in humans were reported [19]. Different from other forms of acquired diaphragmatic dysfunction, VIDD is potentially be treated by simple measures such as sedative regimens selection and ventilator settings [2]. In clinical practice, appropriate anesthetics is essential for maintaining adequate sedation during prolonged MV. A clear understanding of the effects of different anesthetics on the diaphragm is important for the VIDD medical treating strategy. So, a study to compare the effect of midazolam, propofol, and dexmedetomidine on diaphragm function during MV is needed.
Heavy Chain (MY-32), anti-Slow Skeletal Myosin Heavy Chain (NOQ7.5.4D), anti-Laminin were purchased from Abcam (Shanghai, China). SlowFadeTM Gold Antifade Mountant with DAPI and secondary antibodies including Alexa Fluor 546 conjugated donkey anti-mouse, Alexa Fluor 488 donkey anti-goat were purchased from Invitrogen (Shanghai, China). 2.3. Experimental design Two parallel animal experiments were performed in the present study: Experiment 1 (SB group): Animals breathed spontaneously with continuous infusion of pentobarbital (PB) (n = 6), midazolam (MZ) (n = 6), dexmedetomidine (DD) (n = 6) or propofol (PP) (n = 6) for 12 h. Experiment 2 (MV group): Animals were mechanically ventilated with continuous infusion of pentobarbital (PB) (n = 5), midazolam (MZ) (n = 5), dexmedetomidine (DD) (n = 5) or propofol (PP) (n = 5) for 12 h. It has been demonstrated that pentobarbital and the carrier of propofol (Lipovenous 10%) did not affect the diaphragm force [13,17], so we set the pentobarbital sedated groups as the sham. And no more animals were performed to examine the effect of the lipovenous in our study. Besides, the traditional sedation depth evaluation method (foot reflex, corneal reflex, and arterial blood pressure) was used in SB group; a new method (negative foot/corneal reflex, normal arterial blood pressure, peak inspiration pressure (PIP) ≤ 8 cm H2O) was used in MV group. Detail reasons about developing the new sedation depth evaluation method were stated in the appendix. 2.4. Methods All animals underwent a tracheotomy, right carotid artery, and jugular vein cannulation after peritoneal injection of sodium pentobarbital (60 mg•kg−1). Heart rate and mean arterial pressure (MAP) were continuously measured using biologic data acquisition and analysis system (BL-420 F, Wuhan, China). Arterial blood samples were taken every 4 h for blood cell counts (Pentra MS CRP, HORIBA, Japan) and blood gas (i-STAT300 G, Abbott, USA) examination. In addition, venous blood glucose was measured (ONE TOUCH, Johnson, USA) every 4 h at the end of the tail. Body fluid homeostasis was ensured by continuous administrating electrolyte solution (Baxter, Deerfield, IL) through the jugular vein (1 ml•kg−1 h−1). The body temperature was maintained at 37 °C using a temperature controller (BP-2010A, Softron, Japan) in a supine position. Bladder expression, eyes lubrication, and airway mucus removal were operated periodically. The ratio of the airoxygen gas mixture was adjusted to maintain the arterial partial pressures of oxygen (PaO2) between 80 mmHg and 100 mmHg. Animals in the MV group were ventilated using a volume controlled small-animal ventilator (VentElite, Harvard Apparatus, USA). Anesthetics dosages were administered to maintain a relatively steady sedation depth. The respiration waves and PIP were recorded by a connected computer. The tidal volume was set at 6–7 ml•kg−1 body weight with an inspiration/expiration ratio at 1:1. Respiratory rate was 60–70 breaths/minute and adjusted to maintain arterial partial pressures of carbon dioxide (PaCO2) at less than 50 mmHg. No positive endexpiratory pressure was used. The gas mixture was humidified and maintained at 37 °C.
2. Materials and methods 2.1. Animals Specific pathogen-free male Wistar rats weighing about 300 g were purchased from WeiTongLiHua experimental animals Co. Ltd (Beijing, China), and feed in the ABSL-III laboratory of Wuhan University. The study was approved by the Animal Experiment Center of Zhongnan Hospital of Wuhan University (Wuhan, China) with the serial number: 2018039. All procedures were carried out in compliance with the Institutional Animal Care and Use Committee. 2.2. Anesthetics and reagents
2.4.1. Microcirculatory measurements A Sidestream Dark Field (SDF) imaging device (MicroSee, Guangzhou, China) was used to measure the diaphragm microcirculation. The measurement points were set at the time of 30 min (T0), 6 h (T6) and 12 h (T12) after MV initiation. A skin incision 1 cm long was performed via a midline laparotomy to expose the cavity without obvious bleeding. The probe was inserted into the cavity and placed on the surface of the costal and crural diaphragm segments. The
Sodium pentobarbital (Saiyisi, Wuhan, China), dexmedetomidine (Aibeining, Jiangsu, China), propofol 2% (Diprifusor, AstraZeneca, Italy), and midazolam (Liyuexi, Jiangsu, China) were prepared for continuous infusion. Krebs-Henseleit bicarbonate buffer was purchased from M&C Gene Technology Ltd. (Beijing, China). Primary antibodies including anti-atrogin-1, anti-4-HNE, anti-catalase, anti-SOD1/2, antiAkt, anti-p-Akt, anti-Murf-1, anti-HIF-1α, anti-Fast Myosin Skeletal 2
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
manipulation was gently applied without pressure on the diaphragm and liver. Five sequences of 20 s each from different adjacent areas were recorded and stored under an alpha-numeric code. All sequences were acquired by the same manipulator and analyzed by a second blinded investigator. The total vascular density (TVD), microcirculatory flow index (MFI), flow heterogeneity index (HI), and functional capillary density (FCD) were analyzed according to the international recommended guidelines [20]. In addition, the right hindlimb gastrocnemius muscle microcirculation was also measured. Twelve hours later, a strip of costal diaphragm muscle was dissected for in vitro contractile measurements. The remaining diaphragm was stored to further determine biochemical and pathological examination. The lung was also removed and fixed for ventilation induced lung injury (VILI) evaluation.
then separated via SDS-PAGE through a stacking (5%) and separation (8%) gel. Subsequently, proteins were transferred to polyvinylidene fluoride (PVDF) membranes (200 mA for 1 h). The membranes were incubated with primary antibodies at 4 °C overnight. After washing with TBST, the membranes were probed with HRP-labeled secondary antibodies. Then the membranes were detected via chemiluminescence and image densitometry analysis with a luminescent image analysis system (Tanon-6200, China). 2.5. Statistical analysis Data are shown as number or mean ± standard (SD). Normal distribution was assessed with the Shapiro-Wilk test. The primary outcome was the in vitro total tension of the diaphragm and gastrocnemius. The 2-tailed Student’s t test was used for single biochemical or physiological observation comparison. Post-hoc analysis was performed as described if statistical significance (p ≤ 0.05) was achieved. Multiple responses were analyzed by one-way or two-way ANOVA. The relationship between MAP and muscle microcirculation was investigated using linear regression and Spearman correlation. Calculations were performed using Graphpad Prism 6.0.
2.4.2. Diaphragm contractility The method was elaborated in our prior study [9]. A system (ALCMPA2000, Shanghai, China) including organ bath chambers, transducer, and analysis software was used. The Krebs-Henseleit (K-H) solution was bubbled with a gas mixture of 95% O2-5% CO2 and maintained at 37 °C and pH 7.4. Optimal muscle length (Lo) for peak twitch force was established and the following parameters were performed at Lo: 1) maximum twitch force; 2) maximal tetanic force; 3) force-frequency relationship. The weight mass, length, width, and thickness were measured at Lo. All forces were normalized for CSA.
3. Results 3.1. General characteristics of animals in all groups
2.4.3. Immunofluorescence (IF) staining Pieces embedded in Optimal Cutting Temperature (OCT) were cut into 8-μm slices under −20 °C using a frozen slicer (CM1900, Leica, Germany). Two adjacent sections were stained with either mouse antislow myosin heavy chain antibody or anti-fast myosin heavy chain antibody. Additionally, rabbit anti-laminin antibody was included in each of these stains. After reaction with a mixture of Alexa Fluor 546 conjugated donkey anti-mouse and Alexa Fluor 488 donkey anti-goat secondary antibody, the sections were sealed with SlowFadeTM Gold Antifade Mountant with DAPI. Images were obtained using an OLYMPUS IX73 microscope (Olympus Co., Japan) and CSA was calculated with ImageJ software (Fiji) with at least 200 fibers per animal.
At the end of the experiments, MAP was significantly lower in comparison to baseline in all groups. Data of MAP, heart rate, blood gas, lactate, and glucose before euthanization were not significantly differences in all the MV subgroups (p > 0.05, respectively, Table 1) and the SB subgroups (p > 0.05, respectively, Table S1). In addition, differences in the blood cell counts were also insignificant in subgroups of MV and SB. Totally, no obvious glucose disorder, acid-base imbalance and systemic infections in animals in the present study. Anesthetics consumption dosages were significantly lower in the SB group compared to the MV group (p < 0.05, respectively). Table 1 Characteristics of the arterial blood gas, vital signs, blood cell counts, and total sedatives dosages used at the end of experiments in MV group.
2.4.4. Hematoxylin and eosin (HE) staining Tissues fixed with 4% paraformaldehyde were chopped into 0.5 cm × 0.5 cm × 1 cm pieces. Samples underwent dehydration, wax immersion and were then embedded in paraffin (ASP200S, Leica, Germany) for cutting into slices (RM2245, Leica, Germany). 5 μm sections were routinely stained with HE (Baso, Wuhan, China). Images were obtained using an OLYMPUS IX73 microscope (Olympus Co., Japan).
Group Arterial blood gas pH PaCO2 (mmHg) PaO2 (mmHg) Lactate (mmol/L) HCO3− (mmol/L) BEecf (mmol/L) Vital signs Heart rate (beats/ min) Rectal temperature (°C) MAP (mmHg) Glucose (mmol/ L) Blood cell counts Erythrocytes (x1012/L) Hemoglobin (g/ L) Neutrophils (%) Lymphocytes (%) Sedatives dosages (mg)
2.4.5. Transmission electron microscope (TEM) Tissues fixed with 2.5% glutaraldehyde (Servicebio, China) were chopped into 0.1 cm2 × 1 cm pieces. Then the tissues were rinsed, dehydrated, embedded, sliced, stained, and subsequently investigated under a transmission electron microscope (Tecnai G2 20 TWIN, FEI, USA). 2.4.6. Western blotting The protein level of 4-hydroxynonenal (4-HNE), hypoxia inducible factor-1α (HIF-1α), catalase, superoxide dismutase 1/2 (SOD1/2), protein kinase B (Akt), muscle ring finger-1 (Murf-1), atrogin-1, LC3BII/I were measured by SDS-PAGE and immunoblotting as previously described [9]. Briefly, 30 mg of diaphragm samples were homogenized in 200 μl RIPA lysis buffer with cocktail tablets (Servicebio, China) and centrifuged at 1200 RCF units for 15 min at 4 °C. The supernatant was then collected and its protein concentration was determined using the Total Protein Assay kit (BCA method, Beyotime Biotechnology, China). Eighty micrograms of protein from each of the diaphragm samples were
PB (n = 5)
MZ (n = 5)
DD (n = 5)
PP (n = 5)
7.42 ± 0.07 34 ± 4 104 ± 11 1.3 ± 0.2 25.6 ± 2.8 3±1
7.44 ± 0.05 33 ± 3 106 ± 8 1.5 ± 0.3 26.6 ± 2.1 3±1
7.43 ± 0.04 33 ± 4 110 ± 10 1.4 ± 0.3 24.6 ± 2.5 4±1
7.43 ± 0.04 35 ± 4 107 ± 7 1.1 ± 0.2 25.8 ± 2.6 3±2
386 ± 19
390 ± 12
394 ± 14
382 ± 14
37.3 ± 0.2
37.1 ± 0.1
37.0 ± 0.1
37.3 ± 0.3
103 ± 5 5.2 ± 0.8
105 ± 4 5.0 ± 0.7
107 ± 4 5.6 ± 0.9
104 ± 5 5.4 ± 0.4
7.1 ± 0.3
7.5 ± 0.8
6.9 ± 0.4
7.0 ± 0.3
161 ± 8
158 ± 9
165 ± 6
160 ± 6
46.1 ± 3.1 36.9 ± 3.7 47 ± 3
44.8 ± 2.6 35.3 ± 4.0 18 ± 3
44.1 ± 1.7 37.4 ± 3.9 0.106 ± 0.005
45.2 ± 2.3 36.7 ± 4.2 165 ± 3
Data are expressed as number or mean ± SD; PB = pentobarbital group; DD = dexmedetomidine group; PP = propofol group; MZ = midazolam group. BEecf = base excess of extracellular fluid; MAP = mean arterial pressure. 3
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
Fig. 1. Force frequency curves and cross sectional areas of diaphragm in SB and MV groups (mean ± SD). Diaphragm force-frequency curve in SB group (a) and MV group (b); cross sectional areas (CSA) of diaphragm in SB group (c) and MV group (d). Groups represent: PB = pentobarbital group; DD = dexmedetomidine group; PP = propofol group; MZ = midazolam group; white box = type I fibers; grey box = type II fibers. * p < 0.05 vs. pentobarbital group, # p < 0.05 vs. midazolam group.
3.2. Diaphragm contractile properties and cross-sectional areas
3.3. Diaphragm microcirculation in the MV group
Twelve hours of MV or SB resulted in a significant reduction in diaphragm specific force in MZ, DD, and PP groups compared to PB. The discrimination between MZ, DD, and PP were not significant (p > 0.05, respectively) under the SB condition (Fig. 1a). However, midazolam more clearly induced force reduction compared to propofol (−58.5 ± 10.2%, p= 0.0001) and dexmedetomidine (−64.0 ± 10.5%, p = 0.0001) after MV at stimulation frequency ranging from 10 to 120 Hz (Fig. 1b). Although the diaphragmatic forces were impaired, midazolam, dexmedetomidine, and propofol did not induce further diaphragm atrophy compared to pentobarbital under SB condition (Figs. 1c, S1). Dexmedetomidine and propofol were more effective in protecting the fibers from atrophy after MV compared to midazolam, especially in the Type II fibers (p = 0.0001, respectively) (Figs. 1d and 2 ). Different from diaphragm, fiber size and contraction force in gastrocnemius were kept unchanged in MZ, DD, and PP groups compared to PB after12 h of MV or SB (Fig. S2-S4).
No statistical differences of PIP and MAP existed between all the MV groups during the sequences manipulation (Table 2). During the whole process of the measurements (T0, T6, T12), red cells moved fluently in the vessels (diameter ≤ 20 μm), and no obvious of heterogenetic blood flow and vessel occlusion (10 μm ≤ diameter ≤ 20 μm) were observed despite the diaphragm disused (Video S1). There were no significant differences of TVD, MFI, and HI between all the groups at the time of T0, T6, and T12. But a time-dependent decline in FCD was observed in all the groups (Fig. 3a–c). FCD decreased significantly (PB −25.1%, p = 0.0001, MZ −21.6%, p = 0.0003, DD −15.2%, p = 0.022, PP −24.8%, p = 0.0001, respectively) from T0 to T12 (Fig. 3d). Dexmedetomidine was more efficient in preserving the perfused capillary than midazolam or propofol, especially at T12 (p = 0.006, relative to propofol). Weak relationship between FCD and MAP was observed from T0 to T6 (R2 = 0.263, p < 0.001) but not T6 to T12 (R2 = 0.096, p = 0.016) (Fig. S5). Unlike the diaphragm, alterations of microcirculation of the diaphragm were not observed in the gastrocnemius muscle.
Fig. 2. Representative immunofluorescent staining of type I (above) and type II (below) fibers in diaphragm in MV group. Groups represent: PB = pentobarbital group; DD = dexmedetomidine group; PP = propofol group; MZ = midazolam group. 4
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
using midazolam (4.5 ± 0.6%, p = 0.0001) or dexmedetomidine (4.7 ± 0.7%, p = 0.0001) induced higher diaphragmatic autophagic activity than propofol (1.9 ± 0.3%), and these alterations were also presented in the autophagosome formation (Fig. 6).
Table 2 Characteristics of the peak inspiration pressure, mean arterial blood pressure at T0, T6 and T12. Group PIP (cmH2O) T0 T6 T12 MAP (mmHg) T0 T6 T12
PB (n = 5)
MZ (n = 5)
DD (n = 5)
PP (n = 5)
5.4 ± 1.6 5.2 ± 1.3 5.6 ± 1.8
5.8 ± 1.8 5.5 ± 1.4 5.3 ± 1.2
6.1 ± 2.0 5.1 ± 1.5 5.5 ± 1.4
5.4 ± 1.2 5.4 ± 1.6 5.6 ± 1.1
132.4 ± 5 105 ± 4 * 103 ± 5 *
130.1 ± 6 107 ± 5 * 105 ± 4 *
137.2 ± 5 101 ± 4 * 107 ± 4 *
134.8 ± 4 105 ± 4 * 104 ± 4 *
4. Discussion Our experiments provide important information regarding the role that anesthetics play in VIDD. This research systematically compares the effects of dexmedetomidine, propofol, and midazolam on diaphragm function under MV and SB in rats. Conclusively, midazolam, dexmedetomidine, and propofol all have direct adverse effects on the diaphragm contraction force under SB. MV sedated with midazolam results in a more severe diaphragm dysfunction than dexmedetomidine or propofol, possibly caused by its relative weak capacity in inhibiting the diaphragmatic ROS production or protein degradation. Propofol induced similar diaphragm dysfunction compared to dexmedetomidine. MV results in a time-dependent decrease of FCD in the disused diaphragm, and dexmedetomidine is more effective in preserving perfused capillary density than midazolam or propofol.
Data are expressed as number or mean ± SD. PB = pentobarbital group; DD = dexmedetomidine group; PP = propofol group; MZ = midazolam group. PIP = peak inspiration pressure; T0, T6, T12 = 30 min, 6 h, 12 h after mechanical ventilation initiation. * p < 0.05 vs. T0.
3.4. Oxidative stress in the MV group The lipid peroxidation adducts 4-HNE and HIF-1α levels were significantly lower in DD and PP compared to MZ (Fig. 4a-d) (p < 0.05, respectively). Meanwhile, the catalase (Fig. 4e) and SOD levels (Fig. 4f) were also relative lower (p < 0.05, respectively) in MZ compared to DD or PP. Midazolam did not inhibit or eliminate the ROS production in the diaphragm efficiently than dexmedetomidine and propofol.
4.1. Midazolam causes a more severe diaphragm dysfunction and oxidative stress than propofol or dexmedetomidine in ventilated animals As stated above, MV using midazolam sedation induced a more severe diaphragmatic atrophy and contractile dysfunction than dexmedetomidine and propofol. In prior experiments under pentobarbital anesthesia, oxidative stress was recognized as the main factor modifying up or down pathways involved in the process of VIDD [2]. We find dexmedetomidine and propofol but not midazolam significantly decreased the diaphragmatic ROS production compared to pentobarbital. Anesthetics protecting organisms by defending oxidative stress injury have already been reported in various models. Dexmedetomidine reduced oxidative stress caused by ischemia-reperfusion injury in skeletal muscle [14]. Propofol enhanced heme oxygenase expression and generated antioxidant activities because its molecular structure is
3.5. Protein degradation in MV group Regulator of the autophagy-lysosome pathway and the ubiquitin proteasome system-pAkt/Akt ratio was only significantly decreased in DD (−48.1%, p = 0.0001) compared to PB (Fig. 5a, b). Meanwhile, markers that acted as the indicators of ubiquitin proteasome system Murf-1 (Fig. 5c) and atrogin-1 (Fig. 5d) were decreased in DD and PP but not MZ compared to PB. Autophagic activity, represented by the ratio of LC3B-II/LC3B-I, were statistically increased in all the three groups compared to PB (p < 0.001, respectively) (Fig. 5e, f). Sedation
Fig. 3. Representative SDF images and summary of functional capillary density of rat diaphragm microcirculation. Images sequenced at 30 min (a), 6 h (b), and 12 h (c) after MV initiation in pentobarbital group (objective 5x, on screen 325x). Summary of functional capillary density of all MV groups (d). Values are displayed as mean ± SD. FCD = functional capillary density; T0, T6, T12 = 30 min, 6 h, 12 h after MV initiation. Groups represent: PB = pentobarbital group; DD = dexmedetomidine group; PP = propofol group; MZ = midazolam group. * p < 0.05 vs. T0.
5
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
Fig. 4. Assessment of oxidant and anti-oxidant indicators in the diaphragm of the MV group. Representative immunoblots of 4-hydroxynonenal (4-HNE) (a). Ratio of 4-HNE (b) (mean ± SD). Representative immunoblots of hypoxia induced factor-1α (HIF-1α), catalase, and superoxide dismutase (SOD1/2) (c). Ratio of HIF-1α (d), catalase (e), and SOD1/2 (f) (mean ± SD). Groups represent: PB = pentobarbital group; MZ = midazolam group; DD = dexmedetomidine group; PP = propofol group; GADPH = glycerinaldehyd-3-phosphat-dehydrogenase. * p < 0.05 vs. pentobarbital group, # p < 0.05 vs. midazolam group.
similar to vitamin E [21]. Midazolam prevented cells death from oxidative stress mediated by JNK-ERK pathway [12]. From our data, it is concluded that dexmedetomidine and propofol are superior to midazolam in inhibiting lipid peroxidation and increasing antioxidant enzymes activity in the disused diaphragm. And this finding is also supported by the fact that propofol and dexmedetomidine exhibit a superior antioxidant function to midazolam in humans [22]. Midazolam does not, at least not effectively exhibit its antioxidant property in ventilated diaphragm when compared to pentobarbital at sedation dosage. Our results demonstrate that the effects of anesthetics on diaphragm function during MV are associated with their antioxidant properties. Ubiquitin proteasome system activated by the increased oxidative stress was recognized as the main reason causing diaphragm fiber atrophy [23]. Consistent to the ROS level, Murf-1 and atrogin-1 expression levels were decreased differently in DD and PP but not MZ compared to PB. But the ratios of LC3B-II/LC3B-I are statistically increased in all the three groups compared to PB, especially in MZ and DD. Dexmedetomidine increased autophagy activity in VIDD was already been demonstrated [17]. Midazolam was also implied that could induce autophagy to control cell apoptosis [24]. Propofol is most efficient in inhibiting the diaphragmatic autophagic activity and this is in line with other studies [25]. In fact, the function of autophagy in the process of VIDD is still in controversial. It was reported that autophagy was necessary for VIDD, and increased autophagy stimulated by oxidative stress will further resulting in the increase of autophagy and ROS [26]. However, the other study indicated that autophagy induced by MV might instead be a beneficial adaptive response and was not responsible for diaphragm atrophy [27]. Our experiments provide evidence that autophagy is indeed involved in the development of VIDD, and anesthetics have different influences on the autophagy. But
whether these influences are beneficial or adverse for the diaphragm during VIDD is still unclear. 4.2. Dexmedetomidine induced diaphragm dysfunction but prevent muscle atrophy in a low level of oxidative stress and proteolytic activity under MV Dexmedetomidine was already shown to reequilibrate the oxidantantioxidant balance especially by enhancing antioxidant defense enzymes and inhibiting lipid peroxidation in various models [14,28,29]. Our data corroborate these studies. Currently, dexmedetomidine protects the diaphragm from MV is associated with its antioxidant function. Thomas et al concluded that dexmedetomidine led to a worsening of VIDD but diaphragmatic fiber atrophy was prevented even in the circumstance of increased oxidative stress and activated proteolytic pathways [17]. Contrarily, the level of 4-HNE, catalase, and SOD1 in the DD were significant lower compared with PB in our experiment. Thomas et al used a traditional sedation depth evaluation method in their experiments, which led to the prevalence of RVD during MV. We also demonstrated that the diaphragmatic oxidative stress level was significant higher in RVD positive rats compared to the negative ones after 12 h of MV under pentobarbital sedation. Their study could not exclude the extra oxidative stress produced by RVD when ventilation was carried out using traditional sedation depth method. Noticeably, the duration of MV in our experiment is half that of their study. VIDD occurred in 12 h of MV was proved by others [26], and our previous experiment [9]. Whether the additional 12 h’ MV with RVD induce a severer diaphragm dysfunction or higher oxidative stress level by dexmedetomidine sedation is still unknown. Our results indicate that, at least in 12 h, the antioxidant function of dexmedetomidine is reserved in VIDD, and more effective than propofol or midazolam.
6
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
Fig. 5. Assessment of proteasome and autophagy indicators in the diaphragm of the MV group. Representative immunoblots of phosphorylated/total Akt, Murf-1, and Atrogin-1 (a). Ratio of p-Akt/Akt (b), Murf-1 (c), and Atrogin-1 (d) (mean ± SD). Representative immunoblots of of LC3B-II/LC3B-I (e). Ratio of LC3B-II/LC3B-I (f) (mean ± SD). Groups represent: PB = pentobarbital group; MZ = midazolam group; DD = dexmedetomidine group; PP = propofol group; GADPH = glycerinaldehyd-3-phosphat-dehydrogenase. * p < 0.05 vs. pentobarbital group, # p < 0.05 vs. midazolam group.
4.3. Propofol induced similar diaphragm dysfunction compared to dexmedetomidine
Because it was already recognized that pentobarbital did not affect the diaphragm force, the conclusion that propofol did not protect against ventilator-induced diaphragmatic atrophy and the oxidative injury was quite not convincing without a “control” group- MV + pentobarbital. The authors neglected the fact that MV was independent of sedatives contributed to diaphragmatic atrophy and contractile dysfunction. Second, the duration of SB or MV in our experiment is half that of the prior study. Third, the mean propofol infusion dosages in our experiments (27 mg•kg−1 h−1 in SB group, 46 mg•kg−1 h−1 in MV group) are higher than the study used (20 mg•kg−1 h−1). But none of the animals did show any signs of propofol infusion syndrome at our dosages. Moreover, it was reported that a high dose (50 mg•kg−1 h−1) propofol was more beneficial in protecting the diaphragm from sepsis [11] or healthy rats [30]. Fourth, they also used the traditional sedation depth evaluation method which RVD could not be excluded in the MV group. Future study about the effects of different dosages and duration of propofol used on diaphragm function is needed.
Our data show that the diaphragmatic forces and fibers dimensions in PP are similar to DD. Likewise to dexmedetomidine, propofol is also effective in maintaining the diaphragmatic oxidant-antioxidant balance during MV. Our experiment further corroborates the hypothesis that diaphragm dysfunction induced by dexmedetomidine was similar to propofol [17]. But in an experimental model of 24 h of SB, propofol induced diaphragm atrophy directly without obvious oxidative stress injury [12]. This study also concluded that propofol did not protect against ventilator-induced diaphragmatic atrophy and oxidative injury. Differences of the experimental designs and anesthetics dosages used may be helpful to explain the contrary results. First, the set of “control” group is different. We set the animals breathed spontaneously or ventilated mechanically with continuous pentobarbital infusion which is much different from the direct euthanized after anesthetization.
Fig. 6. Representative transmission electrode microscope in diaphragm. Arrow = autophagosome. Groups represent: PB = pentobarbital group; MZ = midazolam group; DD = dexmedetomidine group; PP = propofol group. 7
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
4.4. MV induced diaphragm blood flow reduction most occurred in capillary
a lower antioxidant activity, higher lipid peroxidation and protein ubiquitination level. Sedation with midazolam worsens diaphragm function than dexmedetomidine and propofol during MV. Propofol and dexmedetomidine induced similar diaphragm dysfunction in a low level of oxidative stress and proteolytic activity under MV. MV induced diaphragm blood flow reduction, and this most occurred at the capillary level.
Prior to our study, it was already revealed that 6 h of MV resulted in a 75% reduction in whole blood flow in the disused diaphragm [31]. Our previous study also demonstrated that 12 h of MV resulted in a clear diaphragm blood flow reduction using the laser Doppler technique [9]. Now, we find the reduction of diaphragm blood flow during MV mostly occurred at the capillary using the SDF technique. And this finding is supported by the rapid declined perfused capillary density in the disused diaphragm. The components of the microvasculature (arterioles, capillaries, venules) are able to adapt to the function changed in the disused skeletal muscle. Structure of larger blood vessels (e.g., arterioles) appears to be preserved, and the major adaptation to disuse occurred at the capillary bed level [32,33]. Our results are kept in line with these studies. It was recognized that the microvascular circulation of the tissue was primarily dependent on the perfusion pressure of the organ. Although MAP was decreased during MV, however, FCD continued decreased even MAP was kept stable from T6 to T12. In addition, no relations were observed between MAP and FCD in all groups from T6 to T12. Conclusively, MV induced diaphragmatic FCD reduction was independent to the decrease of MAP. Although the diaphragmatic functional capillary density declined during ventilation, it possibly not the key factor contributed to the ventilator-induced diaphragm dysfunction.
Funding This work was supported by the Emergency Diagnostic and Therapeutic Center of Central China (Wuhan, Hubei, China). Declaration of Competing Interest The authors declare that there are no conflicts of interest. Acknowledgments Thank Yu-Qian Zhao and Qi Li (Intensive Care Unit of Zhongnan Hospital, Wuhan, China) for their help for Microsee (Guangzhou, China) manipulation. Thank Li Yang, Xiao-Hua Leng (Medical Research Center of Zhongnan Hospital, Wuhan, China) for their technique assistance. Thank Di Hu (MAZARS, Wuhan, China) for her help for data analysis.
5. Conclusions Compared to dexmedetomidine or propofol, midazolam resulted in A
Rat-Ventilator Dysyynchrony existed using a traditional sedation depth evaluation method In animals’ experimental models, a traditional method (foot reflex, corneal reflex and arterial blood pressure) was widely used for the sedation depths evaluation. For example, rats with negative foot/corneal reflex and normal arterial blood pressure were recognized as “adequate” sedated to receive prolonged mechanical ventilation (MV) [13,16]. However, during our pre-experiments of MV with four anesthetics sedation, we found ratventilator dyssynchrony (RVD) (evaluated by widely changed airway pressure, Video A1) existed through the whole experiments and could not be improved by ventilator settings adaption when rats’ sedation depth was evaluated by “adequate” level. We found RVD could be avoided using a new sedation depth evaluation method (negative foot/corneal reflex, normal arterial blood pressure, peak inspiration pressure (PIP) ≤ 8 cm H2O) with a continuous PIP and waveforms displayed ventilator (VentElite, Harvard Apparatus, USA). RVD results in a worsen diaphragmatic dysfunction and lung injury To examine the effect of RVD on diaphragmatic structural and contractile properties, we set a RVD group: rats (n = 5) experienced 12 h MV with continuous pentobarbital sedation using traditional sedation depth evaluation method. Subgroup PB in the MV group as the control (Con). Data of
Fig. A1. Representative diaphragmatic HE (left), diaphragmatic sirius red (middle) and lung HE (right) staining. Groups represent: Con = pentobarbital group; RVD = rat ventilator dyssynchrony group. 8
Biomedicine & Pharmacotherapy 121 (2020) 109405
S.-P. Li, et al.
Fig. A2. Lipid peroxidation (4-hydroxynonenal, 4-HNE) in the diaphragm. Groups represent: Con = pentobarbital group; RVD = rat ventilator dyssynchrony group.
MAP, blood gas, lactate, and glucose before euthanization were not significantly different in the two groups (p > 0.05, respectively). There was also no obvious collagen rupture and inflammatory injury existed in both groups (Fig. A1). But RVD induced severe lung injury included more inflammatory cells and erythrocytes in alveolar walls/spaces, and more presence of ruptured and thickened alveolar walls than control. Meanwhile, RVD induced more clearly diaphragmatic contraction force reduction compared to Con (−41.4 ± 6.4%, p = 0.0001) at stimulation frequency ranging from 10 to 120 Hz. However, no significant differences of CSAs existed between the two groups (p = 0.401, respectively,). The 4-HNE level was also higher in RVD group compared to control (p = 0.003, respectively, Fig. A2). Appendix B. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.biopha.2019.109405.
[17] T. Breuer, C. Bleilevens, R. Rossaint, et al., Dexmedetomidine impairs diaphragm function and increases oxidative stress but does not aggravate diaphragmatic atrophy in mechanically ventilated rats, Anesthesiology 128 (4) (2018) 784–795. [18] E. Narimatsu, M. Aimono, Y. Nakayama, et al., The effects of midazolam and ketamine on D-tubocurarine-induced twitch depression in septic rat diaphragm, Res. Commun. Mol. Pathol. Pharmacol. 108 (5–6) (2000) 369–380. [19] H. Roze, A. Germain, V. Perrier, et al., Effect of flumazenil on diaphragm electrical activation during weaning from mechanical ventilation after acute respiratory distress syndrome, Br. J. Anaesth. 114 (2) (2015) 269–275. [20] C. Ince, E.C. Boerma, M. Cecconi, et al., Second consensus on the assessment of sublingual microcirculation in critically ill patients: results from a task force of the European Society of Intensive Care Medicine, Intensive Care Med. 44 (3) (2018) 281–299. [21] C. Liang, J. Cang, H. Wang, et al., Propofol attenuates cerebral ischemia/reperfusion injury partially using heme oxygenase-1, J. Neurosurg. Anesthesiol. 25 (3) (2013) 311–316. [22] C. Han, W. Ding, W. Jiang, et al., A comparison of the effects of midazolam, propofol and dexmedetomidine on the antioxidant system: a randomized trial, Exp. Ther. Med. 9 (6) (2015) 2293–2298. [23] S. Levine, T. Nguyen, N. Taylor, et al., Rapid disuse atrophy of diaphragm fibers in mechanically ventilated humans, N. Engl. J. Med. 358 (13) (2008) 1327–1335. [24] E.C. So, Y.C. Chen, S.C. Wang, et al., Midazolam regulated caspase pathway, endoplasmic reticulum stress, autophagy, and cell cycle to induce apoptosis in MA-10 mouse Leydig tumor cells, Onco. Ther. 9 (2016) 2519–2533. [25] H. Qiao, Y. Li, Z. Xu, et al., Propofol affects neurodegeneration and neurogenesis by regulation of autophagy via effects on intracellular calcium homeostasis, Anesthesiology 127 (3) (2017) 490–501. [26] A.J. Smuder, K.J. Sollanek, W.B. Nelson, et al., Crosstalk between autophagy and oxidative stress regulates proteolysis in the diaphragm during mechanical ventilation, Free Radic. Biol. Med. 115 (2018) 179–190. [27] I. Azuelos, B. Jung, M. Picard, et al., Relationship between autophagy and ventilator-induced diaphragmatic dysfunction, Anesthesiology 122 (6) (2015) 1349–1361. [28] H. Akpınar, M. Nazıroğlu, İ. Övey, et al., The neuroprotective action of dexmedetomidine on apoptosis, calcium entry and oxidative stress in cerebral ischemia-induced rats: contribution of TRPM2 and TRPV1 channels, Sci. Rep. 6 (2016) 37196. [29] V. Sen, A. Güzel, H.S. Şen, et al., Preventive effects of dexmedetomidine on the liver in a rat model of acid-induced acute lung injury, Biomed Res. Int. 2014 (2014) 621827. [30] S. Ulker, D. Tok, S. Kosay, et al., Effects of ketamine, thiopental sodium and propofol on muscle contractures in rat diaphragm in vitro, Int. J. Exp. Pathol. 72 (5) (1991) 527–532. [31] D.R.T. 3rd1, C.S. Bruells, J.N. Stabley, et al., Mechanical ventilation reduces rat diaphragm blood flow and impairs oxygen delivery and uptake, Crit. Care Med. 40 (10) (2012) 2858–2866. [32] A.B. Borisov, S.K. Huang, C. BM, Remodeling of the vascular bed and progressive loss of capillaries in denervated skeletal muscle, Anat. Rec. 258 (3) (2000) 292–304. [33] K. Tyml, M.-C. O, Structural and functional changes in the microvasculature of disused skeletal muscle, Front Biosci 6 (2001) D45–52.
References [1] A.C. Watson, P.D. Hughes, M. Louise Harris, et al., Measurement of twitch transdiaphragmatic, esophageal, and endotracheal tube pressure with bilateral anterolateral magnetic phrenic nerve stimulation in patients in the intensive care unit, Crit. Care Med. 29 (7) (2001) 1325–1331. [2] M. Dres, E.C. Goligher, L.M.A. Heunks, et al., critical illness-associated diaphragm weakness, Intensive Care Medicine 43 (10) (2017) 1441–1452. [3] E.C. Goligher, M. Dres, E. Fan, et al., Mechanical ventilation-induced diaphragm atrophy strongly impacts clinical outcomes, Am. J. Respir. Crit. Care Med. 197 (2) (2018) 204–213. [4] S.K. Powers, M.B. Hudson, W.B. Nelson, et al., Mitochondria-targeted antioxidants protect against mechanical ventilation-induced diaphragm weakness, Crit. Care Med. 39 (7) (2011) 1749–1759. [5] S.N. Hussain, M. Mofarrahi, I. Sigala, et al., Mechanical ventilation-induced diaphragm disuse in humans triggers autophagy, Am. J. Respir. Crit. Care Med. 182 (11) (2010) 1377–1386. [6] A.J. Smuder, M.B. Hudson, W.B. Nelson, et al., Nuclear factor-κB signaling contributes to mechanical ventilation-induced diaphragm weakness, Crit. Care Med. 40 (3) (2012) 927–934. [7] E. Zhu, S. CS, Ventilator-induced diaphragmatic vascular dysfunction, Crit. Care Med. 40 (10) (2012) 2914–2915. [8] J.M. McClung, D. Van Gammeren, M.A. Whidden, et al., Apocynin attenuates diaphragm oxidative stress and protease activation during prolonged mechanical ventilation, Crit. Care Med. 37 (4) (2009) 1373–1379. [9] X.L. Zhou, X.J. Wei, S.P. Li, et al., Interactions between Cytosolic Phospholipase A2 Activation and Mitochondrial Reactive Oxygen Species Production in the Development of Ventilator-Induced Diaphragm Dysfunction, Oxid. Med. Cell. Longev. 2019 (2019) 2561929. [10] J. Cui, H. Zhao, C. Wang, et al., Dexmedetomidine attenuates oxidative stress induced lung alveolar epithelial cell apoptosis in vitro, Oxid. Med. Cell. Longev. 2015 (2015) 358396. [11] K. Mikawa, K. Nishina, S. Kodama, Obara H, Propofol attenuates diaphragmatic dysfunction induced by septic peritonitis in hamsters, Anesthesiology 94 (4) (2001) 652–660. [12] G.Z. Li, H.L. Tao, C. Zhou, et al., Midazolam prevents motor neuronal death from oxidative stress attack mediated by JNK-ERK pathway, Hum. Cell 31 (1) (2018) 64–71. [13] C.S. Bruells, K. Maes, R. Rossaint, et al., Sedation using propofol induces similar diaphragm dysfunction and atrophy during spontaneous breathing and mechanical ventilation in rats, Anesthesiology 120 (3) (2014) 665–672. [14] A. Amigoni, G. Rizzi, A. Divisic, et al., Effects of propofol on diaphragmatic electrical activity in mechanically ventilated pediatric patients, Intensive Care Med. 41 (10) (2015) 1860–1861. [15] X. Dong, Q. Xing, Y. Li, et al., Dexmedetomidine protects against ischemia-reperfusion injury in rat skeletal muscle, J. Surg. Res. 186 (2014) 240–245. [16] M. Cheng, T. Gao, F. Xi, et al., Dexmedetomidine ameliorates muscle wasting and attenuates the alteration of hypothalamic neuropeptides and inflammation in endotoxemic rats, PLoS One 12 (3) (2017) e0174894.
9