Mantilla, Carlos B.

AbstractBackground and ObjectivesOptimizing perioperative analgesia for patients undergoing major lower‐extremity amputation remains a considerable challenge. The utility of liposomal bupivacaine as a component of peripheral nerve blockade for lower‐extremity amputation is unknown.MethodsWe conducted an observational study comparing three different perioperative analgesic techniques for adults undergoing major lower‐extremity amputation under general anesthesia between 2012 and 2017 at an academic medical center: (1) no regional anesthesia, (2) peripheral nerve blockade with standard bupivacaine, and (3) peripheral nerve blockade with a mixture of standard and liposomal bupivacaine. The primary outcome of cumulative opioid oral morphine milligram equivalent utilization in the first 72 hours postoperatively was compared across groups utilizing multivariable linear regression.ResultsA total of 631 unique anesthetics were included for 578 unique patients, including 416 (66%) without regional anesthesia, 131 (21%) with peripheral nerve blockade with a mixture of standard and liposomal bupivacaine, and 84 (13%) with peripheral nerve blockade with standard bupivacaine alone. Cumulative morphine equivalents were lower in those receiving peripheral nerve blockade with combined standard and liposomal bupivacaine compared with those not receiving regional anesthesia (multiplicative increase 0.67; 95% CI 0.50 to 0.90; P = 0.007). There were no significant differences in opioid utilization between peripheral nerve blockade groups (P = 0.59).ConclusionsPeripheral nerve blockade is associated with reduced opioid requirements after lower‐extremity amputation compared with general anesthesia alone. However, the incorporation of liposomal bupivacaine is not significantly different to blockade employing only standard bupivacaine.

Key points Motor units, comprising a motor neuron and the muscle fibre it innervates, are activated in an orderly fashion to provide varying amounts of force. A unilateral C2 spinal hemisection (C2SH) disrupts predominant excitatory input from medulla, causing cessation of inspiratory‐related diaphragm muscle activity, whereas higher force, non‐ventilatory diaphragm activity persists. In this study, we show a disproportionately larger loss of excitatory glutamatergic innervation to small phrenic motor neurons (PhMNs) following C2SH, as compared with large PhMNs ipsilateral to injury. Our data suggest that there is a dichotomy in the distribution of inspiratory‐related descending excitatory glutamatergic input to small vs. large PhMNs that reflects their differential recruitment. AbstractExcitatory glutamatergic input mediating inspiratory drive to phrenic motor neurons (PhMNs) emanates primarily from the ipsilateral ventrolateral medulla. Unilateral C2 hemisection (C2SH) disrupts this excitatory input, resulting in cessation of inspiratory‐related diaphragm muscle (DIAm) activity. In contrast, after C2SH, higher force, non‐ventilatory DIAm activity persists. Inspiratory behaviours require recruitment of only smaller PhMNs, whereas with more forceful expulsive/straining behaviours, larger PhMNs are recruited. Accordingly, we hypothesize that C2SH primarily disrupts glutamatergic synaptic inputs to smaller PhMNs, whereas glutamatergic synaptic inputs to larger PhMNs are preserved. We examined changes in glutamatergic presynaptic input onto retrogradely labelled PhMNs using immunohistochemistry for VGLUT1 and VGLUT2. We found that 7 days after C2SH there was an ∼60% reduction in glutamatergic inputs to smaller PhMNs compared with an ∼35% reduction at larger PhMNs. These results are consistent with a more pronounced impact of C2SH on inspiratory behaviours of the DIAm, and the preservation of higher force behaviours after C2SH. These results indicate that the source of glutamatergic synaptic input to PhMNs varies depending on motor neuron size and reflects different functional control – perhaps separate central pattern generator and premotor circuits. For smaller PhMNs, the central pattern generator for inspiration is located in the pre‐Bötzinger complex and premotor neurons in the ventrolateral medulla, sending predominantly ipsilateral projections via the dorsolateral funiculus. C2SH disrupts this glutamatergic input. For larger PhMNs, a large proportion of excitatory inputs appear to exist below the C2 level or from contralateral regions of the brainstem and spinal cord.

Autophagy is a homeostatic process for the elimination of damaged, unnecessary proteins or organelles within cells. Our overarching hypothesis is that autophagy contributes to neuromuscular dysfunction, specifically increased NMTF seen with aging. Understanding the mechanisms regulating autophagy flux is essential to discovering therapeutic agents which may mitigate the effects of neuromuscular dysfunction across conditions including old age. In recent studies, we found that in 18 month old mice autophagy in the cervical spinal cord is accelerated and neuromuscular transmission failure (NMTF) in diaphragm muscle‐phrenic nerve preparations is increased. Neurotrophin activity, specifically brain‐derived neurotrophic factor (BDNF) signaling through its high affinity receptor tropomyosin‐related kinase receptor subtype B (TrkB) decreases NMTF. Thus, we predict BDNF/TrkB signaling suppresses autophagy in phrenic motor neurons located in the cervical spinal cord. We hypothesize that inhibiting TrkB kinase activity accelerates autophagy in phrenic motor neurons. TrkB F616A mice at 18 months of age were randomized to receive 1 week treatment with vehicle (n=4) or 1NMPP1 (TrkB kinase inhibitor; n=6). Additionally, 18 month old wild type C57BL/6 mice (n=2) were studied. Whole cervical spinal cord (CSC) was harvested for measurements of autophagy marker protein expression analyzed by Western blotting. Data for the vehicle‐treated TrkB F616A and wild type mice was combined given lack of differences for any measurement. p62, a marker associated with the initiation of autophagy, was significantly decreased in the 1NMPP1‐treated group compared to vehicle‐treated or wild type mice. LC3, a protein marking the autophagosome membrane elongation phase, was not significantly changed across groups indicating that LC3‐tagged autophagosomes proceed to lysosomal degradation and do not accumulate. These results support our hypothesis that inhibiting BDNF/TrkB signaling accelerates autophagy flux in the CSC and reveal trophic effects of BDNF/TrkB signaling in the neuromuscular system. Accelerated autophagy may underlie age‐related neuromuscular dysfunction in a BDNF/TrkB signaling dependent manner. Support or Funding Information Support: NIH R01 AG057052

Glutamatergic (Glu) synaptic inputs at phrenic motor neurons (PhMNs) predominantly comprise descending axons from respiratory centers in the ventrolateral medulla. Descending synaptic inputs provide the drive for inspiratory as well as various other behaviors that result in higher force generation. We recently examined pre‐synaptic Glu terminals (VGLUT1 and VGLUT2‐positive) around the somal membrane of PhMNs grouped according to size using a robust confocal imaging‐based technique and semi‐automated image processing, and found an ~10% higher density of Glu terminals at PhMNs in the lower tertile of somal surface area. In parallel studies, we determined the density of AMPA and NMDA mRNA expression in PhMNs grouped according to size, and found that PhMNs in the lower tertile of somal surface area display ~25% higher density of AMPA (Gria2) and NMDA (Grin1) mRNA expression compared to PhMNs in the upper tertile of somal surface area. Assuming an orderly recruitment of motor units based on their intrinsic electrophysiological properties, PhMNs in the lower tertile of somal surface area are likely recruited first in order to accomplish lower force behaviors necessary for inspiration in normoxic as well as in hypoxic (10% O 2 ) and hypercapnic (5% CO 2 ) conditions. Thus, differences in glutamatergic neurotransmission across PhMNs of increasing size support the orderly recruitment of motor units, promoting recruitment of fatigue resistant units for inspiratory behaviors. In adult male rats, unilateral C2 hemisection (C2SH) abolishes inspiratory‐related diaphragm muscle (DIAm) activity ipsilateral to injury, but DIAm activity during higher force behaviors persists, reflecting the contribution of spared inputs, likely contralateral. Accordingly, we hypothesized that C2SH primarily disrupts Glu synaptic inputs to smaller PhMNs, whereas Glu synaptic inputs to larger PhMNs are preserved. By 7 days following unilateral C2SH, there was a ~45% reduction in ipsilateral Glu synaptic input to PhMNs compared to a ~20% reduction contralaterally. Ipsilateral to C2SH, there was an ~60% reduction in presynaptic Glu terminals at PhMNs in lower tertile of somal surface area compared to ~30% reduction at PhMNs in the upper tertile of size. Contralateral to C2SH, the reduction in presynaptic Glu terminals was similar across PhMNs grouped according to somal surface area. These results are consistent with a more pronounced effect of C2SH on inspiratory behaviors of the DIAm, and the preservation of higher force behaviors after C2SH. These results also indicate that Glu synaptic input to PhMNs varies depending on motor neuron size and likely reflects different motor control, including possibly distinct central pattern generator and premotor circuits, onto PhMNs recruited for motor behaviors with varying levels of force generation. Support or Funding Information NIH Grants R01 HL96750 and HL146114

Brain‐derived neurotrophic factor (BDNF) signaling via its high affinity tropomyosin‐related kinase sub‐type B (TrkB) receptor is widely implicated in regulating glutamatergic synaptic transmission in the context of neuroplasticity. At phrenic motor neurons innervating the main inspiratory muscle in mammals (the diaphragm muscle), glutamate is the main excitatory neurotransmitter and acts primarily via ionotropic AMPA and NMDA receptors. In various regions in the CNS, BDNF/TrkB signaling influences rapid phosphorylation of AMPA and NMDA receptor subunits, supporting a main role in neuroplasticity. BDNF/TrkB signaling also plays an important role across various models of injury and respiratory neuroplasticity. In addition, BDNF/TrkB signaling increases AMPA and NMDA receptor expression in the CNS, but whether this effect is also present in motor neurons is presently unknown. BDNF/TrkB signaling modulates motor neuron excitability, which may reflect changes in glutamatergic receptor expression. In the present study, AMPA and NMDA mRNA expression in retrogradely‐labeled phrenic motor neurons was measured using RNAscope in situ hybridization (ACDBio). Adult male rats were randomly assigned to control (untreated) and 11‐day treatment groups (12 ml/day) with intrathecal infusion of the fusion‐protein TrkB‐Fc or artificial cerebrospinal fluid (aCSF) in the region of the phrenic motor neuron pool. Intrathecal catheter tip position was verified at the terminal experimental at ~C4 segment, and the infusion rate was confirmed to restrict delivery to the cervical spinal cord. Expression of both AMPA and NMDA receptor mRNA was evident as puncta in phrenic motor neurons (identified by cholera toxin sub‐unit B – CTB‐labeling) as well as in other large neurons in the ventral horn. Our analysis of mRNA puncta density was restricted to clearly‐identified, retrogradely (CTB)‐labeled phrenic motor neurons containing the mid‐nuclear region in frozen, thin (10μm) longitudinal sections of the spinal cord. At least 25 phrenic motor neurons were stereologically selected for analysis in each animal. No differences in AMPA or NMDA mRNA density were evident between the untreated control and aCSF treated groups, and these results were combined. Across phrenic motor neurons in the control and aCSF groups, there was ~30% variance in AMPA and NMDA mRNA density within each animal, reflecting heterogeneity across motor neurons of varying size, expected to be recruited for the various levels of force production by the diaphragm muscle. Following TrkB‐Fc infusion, there was evidence of decreased AMPA mRNA density in phrenic motor neurons following but no change in NMDA mRNA density. These results support a role for BDNF/TrkB signaling in modulating glutamatergic receptor expression at phrenic motor neurons, which may thus contribute to the regulation of motor neuron excitability and neuroplasticity following injury or disease. Support or Funding Information NIH grants R01 HL96750 and R01 HL146114

Neuromotor control of diaphragm muscle is dependent on the orderly recruitment of phrenic motor neurons (PhMNs) in a size‐dependent manner. Type S and FR motor units comprise smaller, more easily recruited PhMNs that innervate fatigue resistant, lower force‐producing type I and type IIa muscle fibers. Type FInt and FF motor units, comprise larger, rarely recruited PhMNs innervating the more fatigable, higher force‐producing type IIx and/or IIb muscle fibers. The activity and energy demand upon PhMNs of type S and FR motor units is greater than that of PhMNs from type FInt and FF units, due to the high duty cycle (~35%) of ventilation. Accordingly, activity of type FInt and FF motor units is required for higher force behaviors that are performed intermittently over much shorter durations. These include the expulsive/straining maneuvers during coughing, sneezing, defecation and parturition. The energy demand in both PhMNs and diaphragm muscle fibers is met by mitochondria. To preserve homeostasis, mitochondrial functions are regulated by dynamic, continuous cycles of fusion and fragmentation. We hypothesized that mitochondrial density and mitochondrial fusion will be greater in smaller PhMNs, due to their more frequent activation during ventilation. In six adult (6‐month old) Fischer 344 rats, PhMNs were retrogradely labeled by intrapleural injection of Alexa488‐conjugated CTB. Three days later, cervical transdural infusion of MitoTracker Red was performed every 10 min for 1.5 h under anesthesia. Spinal cord tissue was collected following transcardial perfusion with 4% paraformaldehyde, sectioned longitudinally at 70 μm, and mounted onto slides. Two‐channel sequential confocal z‐stack (0.50 μm step size) mosaics of PhMNs and mitochondria were acquired using a 60x oil immersion objective and analyzed using Elements software (Nikon). The surface areas of PhMNs were measured and divided into tertiles. The lower tertile of PhMNs (comprising type S and FR motor units) had greater mitochondrial volume densities, greater mitochondrial aspect ratios and form factors and greater mean individual mitochondrial volumes than upper tertile PhMNs (comprising type FInt and FF motor units). We further characterized the differences in mRNA expression of mitochondrial dynamics associated proteins (Mfn2 and Drp1) in large vs. small PhMNs using fluorescent in‐situ hybridization techniques. Our results suggest that PhMNs of type S and FR motor units have increased mitochondrial density, supporting their high energy demand to sustain ventilation. Their increased mitochondrial fusion may underlie their resilience to PhMN loss in aging and disease. Support or Funding Information Supported by a National Institutes of Health grants R01‐AG044615

Pro‐inflammatory cytokines, like TNFα, mediate a cellular stress response to inflammation that may target mitochondria. In the present study, we hypothesize that in motor neuron‐like NSC‐34 cells, TNFα induces mitochondrial fragmentation and biogenesis in order to maintain overall O 2 consumption of these cells. NSC‐34 cells were differentiated by serum depletion (~24 h) until neurite structures (>50 μm) were observed. Cells were then exposed to TNFα (20 ng/ml) for 24 h. Mitochondria were labeled using MitoTracker Green and imaged using confocal microscopy. Mitochondrial morphometry was assessed by measuring aspect ratio (ratio of long to short axis) and form factor (perimeter 2 /4p·Area). Expression of mitofusin 2 (Mfn2 – fusion protein) and dynamin related protein 1 (Drp1 – fission) was determined by Western blot. From 3D reconstruction of optical slices of NSC‐34 cells, cell and mitochondrial volumes were determined and mitochondrial volume density calculated. Mitochondrial biogenesis was assessed by measuring mitochondrial DNA, and expression of PGC1α was measured. An Agilent Seahorse XF Analyzer was used to measure cellular O 2 consumption. A quantitative histochemical method was used to measure the maximum velocity of the succinate dehydrogenase (SDH max ) reaction within NSC‐34 cells to determine the functionality of the tri‐carboxylic acid (TCA) cycle and the electron transport chain (ETC). Finally, reactive oxygen species (ROS) formation was measured using MitoSox fluorescence. After exposure to TNFα, aspect ratio and form factor of mitochondria decreased by ~15% compared to untreated controls with consistent changes in Mfn2 and Drp1 expression. After TNFα exposure, mitochondrial DNA expression increased as did mitochondrial volume density, together indicating increased mitochondrial biogenesis. After TNFα exposure, O 2 consumption per mitochondrion (normalized for mitochondrial volume density) decreased. This was consistent with a TNFα‐induced decrease in SDH max . Despite the decrease in O 2 consumption and SDH max per mitochondrion after TNFα exposure, ROS formation increased reflecting mitochondrial dysfunction. We conclude that mitochondrial fragmentation, biogenesis and increased mitochondrial volume density are a homeostatic mechanism to decrease O 2 consumption per mitochondrion thereby damping ROS formation, while maintaining overall O 2 consumption to meet ATP demand. Support or Funding Information Supported by NIH grants AG44615 and HL105355.

Neuromuscular dysfunction is common in old age. A hallmark of aging is the aggregation of damaged cytoplasmic structures, especially in post‐mitotic cells like motor neurons. Autophagy flux may be impaired as a result of deficits in the formation of double membrane autophagosomes, the fusion of autophagosomes with lysosomes, or lysosome‐based degradation. Our overarching hypothesis is that autophagy contributes to neuromuscular dysfunction in old age, and that this effect occurs in motor neurons. We evaluated changes in expression levels of core autophagy proteins necessary for the initiation (Beclin1) and elongation (LC3‐I, LC3‐II, ATG7 and ATG5‐ATG12 complex) phases of autophagsome formation, as well as lysosome degradation (assessed by p62 clearance), in adult male and female C57BL/6 mice at 6, 18, and 24 months of age (reflecting 100%, 90% and 75% survival, respectively). Protein expression was analyzed by Western blotting in the cervical spinal cord segments C2–C6. Expression of p62 was reduced in the cervical spinal cord of adult mice at 18 months compared to 24 months, suggesting increased clearance and autophagy flux at 18 months of age. There was no evidence of an effect of age on the expression of the other autophagy markers. In separate animals, immunohistochemical (IHC) analyses of the C2‐C6 segments was used to determine the relative expression of LC3 and p62 in gray and white matter regions. Across all age groups, LC3 and p62 immunoreactivity was evident in the gray matter with minimal expression in the white matter. The ratio of both LC3 and p62 in motor neurons vs. gray matter was compared using a mixed linear model with animal as a random effect. No effect of sex was evident. Increased expression of both LC3 and p62 was evident in choline acetyl transferase (ChAT)‐positive motor neurons (~2–3 fold) compared to the remainder of the gray matter. There was evidence of a difference in relative LC3 and p62 expression in motor neurons across age (F 349,2 =12.5, p<0.001), with a significant increase by 24 months of age (p<0.005). Taken together, these findings suggest that 1) there is an aging effect on autophagy markers in the cervical spinal cord of mice; 2) an increase in autophagy by 18 months of age reflects an increase in autophagosome clearance (reduced p62 expression) not restricted to motor neurons; and, 3) increased motor neuron expression of both LC3 and p62 by 24 months of age reflect autophagsome accumulation and autophagy impairment in motor neurons. These findings are also is in accordance with previous reports suggesting a period of vulnerability between 18 and 24 months of age in mice, where autophagy may play a role in the development of age‐related neuromuscular dysfunction. Support or Funding Information Supported by NIH R01 AG057052.

The diaphragm (DIA) muscle, which separates the abdominal and thoracic cavities, is the primary inspiratory muscle in mammals. Both transdiaphragmatic pressure (Pdi) and DIA electromyography (EMG) are used to assess respiratory activity across motor behaviors of varying levels of force. However, due to the proximity of the heart to the DIA, cardiac electrical activity can severely impact the quality of the EMG signal, especially in real‐time settings when cardiac activity cannot be removed. Classically, either boxcar functions of root mean square (RMS) type calculations or low‐pass filters are used to quantify DIA EMG activity; however, these techniques have the drawback of being highly influenced by cardiac noise, which occurs approximately four times more frequently than DIA EMG activity during quiet breathing (eupnea). In humans with surface DIA EMG, fixed sample entropy (fSE)—a tool for assessing the complexity of a time series—has been proposed as a method to quantify the intensity of DIA EMG activity across varying levels of muscle pressure generation. The parameters of fSE can be adjusted to be resistant to relatively slow changes in a signal resulting from cardiac activity while being highly susceptible to the relatively quick changes in motor unit action potentials in the time domain. Unsurprisingly, cardiac noise is a problem in non‐human animal studies as well, particularly in disease models when the DIA EMG activity levels are reduced. In the present study, we assessed the effectiveness of fSE in accurately reflecting changes in EMG activity in healthy rats with in‐dwelling DIA EMG electrodes. We hypothesized that fSE EMG would be resistant to changes in the signal caused by cardiac noise while reflecting changes caused by DIA EMG. We evaluated 24 different parameter settings for fSE assessments to optimize the correlation coefficient between the fSE and Pdi signals. In addition, we analyzed the RMS, the average rectified value (ARV), and the fSE of DIA EMG events during eupnea, exposure to an hypercapnic (5% CO 2 )‐hypoxic (10% O 2 ) environment, and during airway occlusion by performing detailed comparisons of individual respiratory events. In total, approximately 900 individual respiratory events per analysis technique were assessed across all subjects. Our results show that fSE EMG is extremely effective at eliminating cardiac noise and has a high correlation coefficient with the Pdi signal (.92). In contrast, the correlation coefficient between RMS EMG and Pdi is 0.83 and the correlation coefficient between ARV EMG and Pdi is 0.88, likely due to the high level of cardiac noise in the DIA EMG. Additionally, we show that ventilatory parameters such as instantaneous burst duration and respiratory frequency are similar between Pdi, RMS, ARV, and fSE EMG. These findings suggest that fSE is a reliable and effective alternative to using traditional RMS and ARV EMG techniques to estimate the magnitude of respiratory muscle electrical activity in muscles that are susceptible to cardiac noise.

There is an age‐associated loss of phrenic motor neurons (PhMNs) and withdrawal of pre‐synaptic terminals at neuromuscular junctions (NMJs) in 24‐month old Fischer 344 (F344) rats. Previously, we showed that age‐related PhMN loss is concomitant with neuromuscular transmission failure and diaphragm muscle (DIAm) sarcopenia (an age‐related atrophy of muscle fibers and a decrease in specific force). It is unknown if pre‐synaptic terminal withdrawal precedes PhMN loss, which would implicate a reduction in a trophic influence emanating from muscle fibers (myogenic). We hypothesized that age‐related PhMN loss results from pre‐synaptic terminal withdrawal and reduced myogenic influence, which occurs predominantly at type IIx and/or IIb DIAm fibers (innervated by larger PhMNs). F344 rats were obtained from the National Institutes of Aging. Experimental ages were chosen to represent 100% and 50% survival at 6 months and 24 months‐old, respectively. PhMNs were first retrogradely labeled by intrapleural injection of cholera toxin‐subunit B (CTB), which is dependent on binding to pre‐synaptic terminal gangliosides. Thus, with pre‐synaptic terminal withdrawal, PhMNs would not be labeled. Subsequently (3 days after CTB), PhMNs were retrogradely labeled by sectioning the phrenic nerve and dipping the proximal end into a Rhodamine dextran (Rho) solution – this labeling is not dependent on pre‐synaptic terminals. CTB and Rho labeled PhMNs were visualized by confocal microscopy. Only optical slices containing the nucleus of Rho‐ and/or CTB‐labeled PhMNs were counted. We analyzed the differences in labeling of PhMNs by CTB and Rho in both age groups. In labeled PhMNs, short and long axis measurements were used to determine somal surface areas by assuming a prolate spheroid shape. In 6‐month old rats, there was no significant difference in the number of PhMNs labeled by CTB or Rho. In 24‐month old rats, the total number of labeled PhMNs was reduced compared to the 6‐month old rats, which primarily affected larger PhMNs. Furthermore, at 24‐months, fewer PhMNs were labeled with CTB compared to Rho and those labeled with Rho had larger somal surface areas. These results support our hypothesis that the age‐related loss of larger PhMNs results from pre‐synaptic terminal withdrawal and reduced myogenic influence. Support or Funding Information Support: NIH R01‐AG044615 American Physiological Society