Search

BackgroundThe lack of physical activity is a common issue in modern society and is considered a major risk factor for various chronic non-communicable diseases. Bioactive factors secreted by skeletal muscle during exercise play a crucial role in inter-organ interactions. Since the concept of “myokines” was proposed in 2004, hundreds of regulatory myokines have been identified. Visual analysis of research on exercise-regulated myokines is significant to explore research hotspots and frontiers in this field.MethodsResearch literature on exercise-regulated myokines from 2003 to 2023 in the “Web of Science” database was used as the data source. Knowledge maps were drawn using “VOS Viewer, CiteSpace, and R-bibliometrix” software.ResultsA total of 1,405 papers were included, showing a fluctuating yet slow growth in annual publications. The United States and China led in the number of publications and collaboration networks. Harvard University ranked first with 120 publications. CIBER (centrality 0.16) and the University of California System (centrality 0.16) were pivotal in advancing this field. PEDERSEN BK led author rankings with 41 publications and 1,952 citations. FRONTIERS IN PHYSIOLOGY ranked first among journals with 64 publications and the highest g-index (39), while PLoS One had the highest h-index (25) and most citations (2,599). Key co-cited reference clusters included #1 skeletal muscle dysfunction, #2 obesity, #6 ASCs, and #7 adaptive immunocytes. Pontus Boström’s paper had a notable citation burst intensity of 77.37. High-frequency keywords were “exercise” (509), “skeletal muscle” (452), and “expression” (293), with long-term keywords such as #0 irisin, #2 insulin resistance, #3 transcription, and #6 physical activity. Recently, keywords like “physical exercise,” “resistance exercise,” “aerobic exercise,” “insulin,” and “oxidative stress” have emerged.ConclusionResearch in the field of exercise-regulated myokines shows an overall upward trend. The focus areas include myokines mediated by different types of exercise, the interaction of irisin-mediated muscle with other organs, and the important role of myokine-mediated oxidative stress in exercise simulation.

Abstract According to the Montreal Protocol, the production and consumption of ozone-layer-depleting CCl4 for dispersive applications was globally phased out by 2010, including China. However, continued CCl4 emissions were disclosed, with the latest CCl4 emissions unknown in eastern China. In the current study, based on the atmospheric measurements of ~12,000 air samples taken at two sites in eastern China, the 2021–2022 CCl4 emissions are quantified as 7.6 ± 1.7 gigagrams per year. This finding indicates that CCl4 emissions continued after being phased out for dispersive uses in 2010. Subsequently, our study identifies potential industrial sources (manufacture of general purpose machinery and manufacture of raw chemical materials, and chemical products) of CCl4 emissions.

Recent studies have shown that clustering latent variables, derived from autoencoders (AEs) through reconstruction tasks, is effective for identifying spatial domains of spatial transcriptomics (ST). Despite their utility, AEs exhibit inherent limitations due to discontinuities in latent spaces where similar inputs might not closely map. Variational Autoencoders address this by enhancing smoothness and continuity through alignment with priors, yet their interpolation can obscure necessary group distinctions. To overcome these limitations, we introduce stBERT, a BERT-based pre-training framework that transforms the traditional reconstruction task in ST into a masked language modeling task. stBERT leverages BERT’s hidden states to achieve refined and rich latent space mappings, preserves essential group distinctions in complex and high-dimensional biological data, to facilitate more effective clustering and interpretation. Our experiments demonstrate that stBERT significantly outperforms current state-of-the-art models in tasks related to spatial domain identification, including clustering validation using ground truth labels, intrinsic evaluation of clustering performance, and biological validation of clustering outcomes. The source code of the stBERT model is available at https://github.com/azusakou/stBERT.

Recent advancements in sequence-structure co-design have demonstrated the efficacy of integrating pre-trained language models (PLMs) with Graph Neural Network (GNN)-based modules for antigen-specific antibody design. A significant advantage of PLMs is their ability to directly predict the Complementarity-Determining Region (CDR), or reduce reliance on structure data by providing a good initialization for the downstream GNN modules. However, while the performance scaling law suggests that training larger models can lead to improved performance, scaling up PLMs for downstream modules faces limitations due to constraints on the size of datasets used for training and fine-tuning. To address the limitations of larger PLMs trained on constrained datasets, we introduce AbGraftBERT. Initially, inspired by the depth up-scaling approach, we provide a Self-Grafting strategy by dividing a PLM into two components: the root layers (first 10 layers, with parameters frozen) and the crown layers (last 10 layers, with parameters frozen), then interconnected these layers through a full-connection layer inspired by the model reprogramming strategy. Furthermore, we introduce Cross-Embedding Attentional Perturbation (CEAP) which perturbs the embeddings passed to the downstream GNN, enabling the model to enhance its focus on critical amino acids and improve robustness. Our experiments demonstrate that our model surpasses current state-of-the-art models in antibody-related tasks, including sequence and structure prediction, antigen-binding antibody design and optimization of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) neutralization. The source code of the AbGraftBERT model is available at https://github.com/azusakou/AbGraftBERT

To assist the low-cost manufacturing of tanks, a one-step method of marginal-restraint mandrel-free spin forming is therefore proposed in this study for the forming of top covers with flanged holes. With the finite element simulation analysis, three different forming strategies are discussed, and the best spinning process is identified. The proposed forming strategies involve preforming the flanged surface first and then preforming the spherical and flanged surfaces in a subsequent trajectory and is proved to have better forming accuracy. Furthermore, the forming quality of the spun parts is improved and optimized by employing more passes, smaller feed ratios, and larger roller fillet radii. It was experimentally verified that the marginal-restraint mandrel-free spinning one-step method with five passes, a feed ratio of 1 mm/r, and a roller fillet radius of 60 mm for the cylindrical roller can achieve accurate forming of a top cover with a center flange hole.

Methyl halides (CH3Cl, CH3Br, and CH3I) are ozone-depleting substances. Biomass burning (BB) is an important source of methyl halides. The temporal variations and global spatial distribution of BB methyl halide emissions are unclear. Thus, global methyl halide emissions from BB during 2003–2021 were estimated based on satellite data. A significant decreasing trend (p

With the increasingly prominent role of social media in the timeliness and sharing of information dissemination, more and more research has focused on how to further improve user stickiness through social media. However, there is little consideration of the impact of celebrities’ views on user behavior in social media. The main goal of this paper is to study the influence of celebrity language style on user communication and opinions dissemination. First, it analyzes the language style characteristics of celebrities’ opinions and conducts cross-influence analysis between celebrity language style characteristics and user communication characteristics. Based on speech act theory, this study studies the influence of different language styles of celebrity Microblog on users’ communication behavior and then builds a potential category analysis model to subdivide the views of celebrities. The results show that (1) Positive expression is the most common language style element combination of celebrities, and it also shows the most effective communication effect. This shows that users like to see celebrities show an active and positive side to the outside world, can analyze external things, and express their own opinions on these contents; (2) The combination of positive emotion, external attention, and analysis can produce the best communication effect; (3) The emotion of celebrities’ opinions will affect the communication emotion of users to a certain extent, and the communication of users will have the development trend of reducing positive emotions and increasing negative emotions. Therefore, positive guidance and the dissemination of positive energy are more needed on public social platforms to minimize or avoid the dissemination of negative emotions.

A thin-walled head with a lateral normal flanged hole is a key part of a propellant tank, and its low-cost and high-precision forming process is very challenging. In this paper, a two-step method is proposed to preform the head via marginal-restraint mandrel-free spinning and then to realize the flanging of the lateral normal hole using a punching–spinning method. Finite element analysis of round-hole punching–spinning flanging shows that the larger the thinning ratio and the roller fillet radius, the lower the rebound and contour rise amount; the larger the feed ratio, the lower the rebound and the higher the contour rise amount. Further study on the effects of round-hole punching–spinning flanging on the secondary deformation of thin-walled heads shows that the deformation of the head in the area around the flanging hole is less severe than that of the flat blank, and the deformation in the circumferential direction is different from that in the busbar direction. Finally, it is verified through experiments that the two-step method can realize the spin forming of thin-walled heads with lateral normal flanged holes.

The special-shaped tank dome of a launch vehicle is a large, thin-walled, curved structure that is difficult to form using the conventional center-restraint spinning method. This study proposes a two-step marginal-restraint mandrel-free spinning method for forming large domes. The finite element analysis results indicate that a larger roller fillet radius and larger feed ratios lead to a larger upper convex angle and the minimum thickness value for the bottom contour. This study explored the impact of shape parameter variations on the upper convexity and transition rounding angle on forming accuracy. The results show that the convexity of the bottom of the special-shaped domes increases with a larger roller fillet radius and larger feed ratios while the overall height decreases. The forming accuracy is adversely affected by larger transition rounding angles and smaller upper convexities. For the accurate forming of domes, the mutual coupling influence during two-step forming should be considered, and a suitable process and suitable trajectory compensation parameters must be carefully selected. Finally, the study verified that a two-step marginal-restraint mandrel-free spinning method with a 10 mm roller fillet radius, a 2 mm/r feed ratio, and the corresponding trajectory compensation can achieve the precise forming of 2250 mm thin-walled special-shaped domes.

Accurate measurement and estimation on the trends and spatial distributions of VOCs emissions in China are critical to establishing efficient local or regional pollution control measures, while less is known about the discrepancies on VOCs emissions estimated by previous studies. In this study, two of the estimation approaches including the bottom-up and top-down methods have been reviewed with the data collected from many studies. The approaches demonstrated that the total anthropogenic VOCs emissions in China have been increasing since 1949. The contributions of industrial and solvent use to total VOCs emissions have been increasing since 2000, whereas the contributions of transportation sector have shown a decreasing trend since 2000. The contributions of fuel combustion have also been decreasing since 1950. The gaps of emission estimates for the industry and solvent use were 99.3 ± 22.7% and 81.5 ± 41.8%, respectively, which distributed in much wider ranges than other sources (e.g. 28.9 ± 16.7% for fuel combustion). In comparison to the top-down method, larger variations on the annual VOCs emission estimates were seen using the bottom-up method that comprised different data sources. For the view of spatial pattern, most hot emission estimate spots were concentrated in the eastern China, consistent to their relatively stronger strengths in the industrialization and urbanization. Although the total VOCs emission in China has been continuously increasing during 2008–2016, the VOCs emissions per gross domestic production (GDP) showed a decreasing trend. As for individual compounds, large discrepancy was seen on formaldehyde, with the coefficient of variation (CV) ranged from 37% to 128% over the years. In overall of view, the importance of industrial process and solvent use is increasing. More focuses must be made to these two sources. Emissions of individual compound, particularly those of oxygenated VOCs, were not completely determined and should be better quantified.

Arthritis is the most common cause to lead to chronic pain. Botulinum toxin type A (BoNT/A) has been widely used to treat chronic pain. In our previous study, we confirmed the anti-inflammatory and antinociceptive effects of BoNT/A in the Complete Freund’s Adjuvant (CFA)-induced arthritis model, but the underlying anti-inflammatory mechanism was not fully elucidated. The purpose of this study was to investigate the anti-inflammatory effects and mechanisms of BoNT/A on arthritis using transcriptomic analysis. The BoNT/A was injected into the rat ankle joint on day 21 after CFA injection. The von Frey and hot plate tests were applied to assess the pain-related behaviors at different time points. Five days after BoNT/A treatment, gene expression profiling in dorsal root ganglion (DRG) was performed using RNA sequencing (RNA-seq). The differentially expressed genes (DEGs) were analyzed by various tools. The mechanical allodynia and thermal hyperalgesia were significantly reversed after BoNT/A injection. RNA-seq revealed 97 DEGs between the CFA group and Sham group; these DEGs were enriched inflammatory response, IL-17 signaling pathway, etc. There are 71 DEGs between the CFA+BoNT/A group and the CFA group; these DEGs related to response to peptide, PI3K-Akt signaling pathway, ECM–receptor interactions, etc. Three key genes were significantly decreased after CFA-induced arthritis pain, while BoNT/A increased the expression of these genes. The identification of S100A9, S100A8, and MMP8 genes can provide new therapeutic targets for arthritis pain and affect the signaling pathway to play an anti-inflammatory role after the treatment of BoNT/A.

Intracerebral hemorrhage (ICH) is a dangerous neurological disease. The mechanism of ferroptosis in ICH remains unclear. Using bioinformatics analysis, we aimed to identify the key molecules involved in ferroptosis and provide treatment targets for ICH to further explore the mechanism of ferroptosis in ICH. GSE24265 was downloaded from the Gene Expression Omnibus (GEO) dataset and intersected with ferroptosis genes. A total of 45 differentially expressed genes (DEGs) were selected, most of which were involved in the TNF signaling pathway and oxidative stress response. Key modules constructed by the protein–protein interaction (PPI) network analysis and screening of genes related to the TNF signaling pathway led to the confirmation of the following genes of interest: MAPK1, MAPK8, TNFAIP3, ATF4, and SLC2A1. Moreover, MAPK1 was one of the key genes related to TNF signaling and oxidative stress, and it may play an important role in ferroptosis after cerebral hemorrhage. The MAPK1-related molecules included hsa-miR-15b-5P, hsa-miR-93-5P, miR-20b-5p, SNHG16, XIST, AC084219.4, RP11-379K17.11, CTC-444N24.11, GS1-358P8.4, CTB-89H12.4, RP4-773N10.5, and FGD5-AS1. We also generated a hemorrhage rat model, which was used to conduct exercise intervention in ICH rats, and qRT-PCR was used to assess the expression levels of our genes of interest. The mRNA levels after cerebral hemorrhage showed that MAPK1, ATF4, SLC2A1, and TNFAIP3 were upregulated, whereas MAPK8 was downregulated. Treadmill training increased the expression of anti-inflammatory molecules TNFAIP3 and SLC2A1 and reduced the expression of MAPK1, ATF4, and MAPK8, indicating that treadmill training may be utilized as antioxidant therapy to decrease neuronal ferroptosis. The results of this study indicated that the MAPK1-related mRNA–miRNA–lncRNA interaction chain could be potentially employed as a biomarker of the inception and progression of ferroptosis after cerebral hemorrhage.

In Japan, the activities of local assembly members in prefectural assembly sessions are crucial within the framework of representative democracy. This study provides a comprehensive review of existing research that quantifies the activities of local assembly members through topic modeling. Furthermore, the BERTopic model was implemented to perform semantic clustering and topic labeling by combining bidirectional encoder representations from transformer (BERT) embeddings and clustering algorithms. The researchers used a dataset of local assembly minutes transformed into a machine-readable format and selected a subset of general questions (159 general questions from 32 members between 2011 and 2015, and 122 general questions from 23 members between 2016 and 2020) in Fukushima Prefecture. The results demonstrate that BERTopic exhibits robust semantic representation and visualization capabilities, thereby improving the understanding of relationships among the main topics. Compared to the Latent Dirichlet Allocation (LDA) and the Structural Topic Model (STM), BERTopic outperformed in terms of identifying subtle topics and yielded a higher topic coherence score, indicating more meaningful and consistent thematic extraction. Furthermore, this research contributes to documenting the recovery process in disaster-affected prefectures after the Great East Japan Earthquake and provides reference material for recovery efforts. [ABSTRACT FROM AUTHOR]

The ultra-high voltage direct current (UHVDC) transmission composed of modular multilevel converters (MMC) is an important technology for large-scale centralized transmission of renewable energy. In UHVDC system, temporary faults in the AC power grid system are a high probability fault, and the fault control strategy affects the safety and reliability of system operation. This paper studies on the condition of AC power grid faults occurring at the receiving converter station. Firstly, study the system characteristics after the occurrence of AC faults, and use theoretical analysis to derive the trend of DC voltage changes of each converter valve. Then, an AC fault ride-through control strategy with high power transmission capability is proposed with hierarchical connection structure, the strategy controls the system to synchronously reduce DC voltage and AC active power after a fault, maximizing the retention of the system’s transmission capacity during the occurrence of faults, thereby reduce power shock in the system. Finally, a simulation model of the dual ended system has built based on the PSCAD simulation platform. The simulation results show that when a single-phase ground fault and a three-phase ground fault occur in the high valve group at the receiving station, the system can retain about 83% and 50% of the transmission capacity during the fault period, respectively. Meanwhile, there is no serious overvoltage or overcurrent phenomenon in the system. The simulation results verified the effectiveness of the proposed control strategy.

The DC negative corona of needle-plate electrodes can generate atmospheric pressure low-temperature plasma active particles, which have important effects on biological mutagenesis. The DC negative corona discharge of an air needle-plate electrode with effective consideration of NOx particles was simulated and the Trichel pulse current was obtained, focusing on the development of particles and the distribution of active nitrogen oxides (RONS) at four moments in the pulse process. The simulation results indicate that the positive ions (N2+ and O2+) and negative ions (O− and O2−) were closely related to the current changes, and the negative ions (O− and O2−) presented a typical stratification phenomenon. RONS (H2O2, O3, and NO) were approximately uniformly distributed above the level of the plate electrode at the same instant, with H2O2 and O3 except for the area below the needle tip. They trended to a cumulative increase in concentration with time. This study provides a theoretical basis for corona discharge plasma seed treatment technology.

Gold chloride clusters play an important role in catalysis and materials chemistry. Due to the diversity of their species and isomers, there is still a dearth of structural studies at the molecular level. In this work, anions of AunCln+3- and AunCln+5- (n = 2–4) clusters were obtained by laser desorption/ionization mass spectrometry (LDI MS), and the most stable isomers of AunCln+3- were determined after a thorough search and optimization at the TPSSh/aug-cc-pVTZ/ECP60MDF level. The results indicate that all isomers with the lowest energy have a planar zigzag skeleton. In each species, there is one Au(III) atom at the edge connected with four Cl atoms, which sets it from the other Au(I) atoms. Four growth pathways for AunCln+3- (n = 2–7) clusters are proposed (labelled R1, R2, R3 and R4). They are all associated with an aurophilic contact and are exothermic. The binding energies tend to stabilize at ∼ −41 kcal/mol when the size of the cluster increases in all pathways. The pathway R1, which connects all the most stable isomers of the respective clusters, is characterized by cluster growth due to aurophilic interactions at the terminal atom of Au(I) in the zigzag chains. In the pathway of R4 involving Au-Au bonding in its initial structures (n ≤ 3), the distance between intermediate gold atoms grows with cluster size, ultimately resulting in the transfer of the intermediate Au-Au bonding into aurophilic interaction. The size effect on the structure and aurophilic interactions of these clusters will be better understood based on these discoveries, potentially providing new insights into the active but elusive chemical species involved in the corresponding catalytic reactions or nanoparticle synthesis processes.

Deep learning, a crucial technique for achieving artificial intelligence (AI), has been successfully applied in many fields. The gradual application of the latest architectures of deep learning in the field of time series forecasting (TSF), such as Transformers, has shown excellent performance and results compared to traditional statistical methods. These applications are widely present in academia and in our daily lives, covering many areas including forecasting electricity consumption in power systems, meteorological rainfall, traffic flow, quantitative trading, risk control in finance, sales operations and price predictions for commercial companies, and pandemic prediction in the medical field. Deep learning-based TSF tasks stand out as one of the most valuable AI scenarios for research, playing an important role in explaining complex real-world phenomena. However, deep learning models still face challenges: they need to deal with the challenge of large-scale data in the information age, achieve longer forecasting ranges, reduce excessively high computational complexity, etc. Therefore, novel methods and more effective solutions are essential. In this paper, we review the latest developments in deep learning for TSF. We begin by introducing the recent development trends in the field of TSF and then propose a new taxonomy from the perspective of deep neural network models, comprehensively covering articles published over the past five years. We also organize commonly used experimental evaluation metrics and datasets. Finally, we point out current issues with the existing solutions and suggest promising future directions in the field of deep learning combined with TSF. This paper is the most comprehensive review related to TSF in recent years and will provide a detailed index for researchers in this field and those who are just starting out.

Realgar- and cinnabar-containing AnGongNiuHuang Pill (AGNHP) is widely used for treating encephalopathy syndrome. However, it raises great safety concerns due to the adverse effects reported by arsenic or mercury poisoning. Although AGNHP has been generally recognized, little is known about the metabolism of arsenic and mercury and their resulting potential health risk in vivo. Thus, comparative pharmacokinetics and urinary excretion of arsenic and mercury were conducted in rats after oral administration of realgar, cinnabar and AGNHP, respectively. The contents of arsenic and mercury in rat blood and urine were determined by hydride-generation atomic fluorescence spectrometry (HG-AFS) after wet digestion. AGNHP significantly reduced the absorption of arsenic in blood and promoted urinary arsenic excretion. Whereas, it increased the blood mercury absorption and reduced urinary mercury excretion. No significant toxicity was observed in the clinical dose range of AGNHP. However, excessive exposure to arsenic and mercury may still pose risks especially by long-term or excessive medication. The results are helpful for the rational clinical applications of realgar- and cinnabar-containing TCMs.

Background Temporal estimation can be influenced by pain, which is a complex psychological and physiological phenomenon. However, the time range in which perception is most sensitive to pain remains unclear. Methods In the present study, we explored the effects of acute inflammatory pain on time perception in the sub- to supra-second (0.6–2.4-s) and supra-second (2–8-s) ranges in rats. Plantar formalin injection was used to induce acute inflammatory pain, and a temporal bisection task was used to measure time perception. Task test sessions were held for five consecutive days (one per day): the day before injection (baseline), immediately after injection, and the three post-injection days. The point of subjective equality (PSE, which reflects the subjective duration) and Weber fraction (which reflects temporal sensitivity) were calculated and analysed. Results In the 0.6–2.4-s range, the PSE was significantly lower, indicating prolonged subjective duration, in the formalin group relative to the saline group (p = 0.049) immediately after injection. Formalin-induced pain also tended to lengthened time perception in the 0.6–2.4-s range on post-injection days 2 (p = 0.06) and 3 (p = 0.054). In the 2–8-s range, formalin injection did not affect the PSE or Weber fraction. Conclusions The enhanced effect of pain on temporal perception in the sub- to supra-second range is observed in this study and this effect is attenuated with the prolongation of estimated time, even in rats.

Abstract The regulation of inflammatory responses by innate immune receptors is recognized as a crucial step in the development of atherosclerosis, although the precise molecular mechanisms remain to be elucidated. This study focused on illustrating the roles of dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN)- and Toll-like receptor 4 (TLR4)-regulated inflammatory responses in macrophages. We found that DC-SIGN expression levels were increased in macrophages of atherosclerotic plaques. Oxidized low-density lipoprotein (oxLDL) significantly enhanced DC-SIGN protein expression levels after a short-term exposure. Knockdown of DC-SIGN decreased expression and secretion of interleukin 1-β (IL1-β), monocyte chemo-attractant protein 1 (MCP-1), tumor necrosis factor-α (TNFα) and matrix metalloproteinase-9 (MMP-9). Immunofluorescence studies demonstrated that DC-SIGN and TLR4 co-localized in regions of the plaques. Moreover, DC-SIGN was co-expressed with TLR4 on the plasma membrane after oxLDL stimulation. The presence of an endogenous interaction and the results of the in vitro pull-down assays revealed that DC-SIGN binds directly with TLR4. We also present evidence that DC-SIGN mediates TLR4-regulated NFκB activation but not activation of p38 and JNK. Our results suggest an essential role of DC-SIGN/TLR4 signaling in macrophages in the pathogenesis of atherosclerosis.

The therapeutic goals of patients with chronic pain are not only to relieve pain but also to improve the quality of life. Chronic pain negatively affects various aspects of daily life, such as by decreasing the motivation to work and reward sensitivity, which may lead to difficulties in daily life or even unemployment. Human and animal studies have shown that chronic pain damages reward processing; the exploration of associated internal mechanisms may aid the development of treatments to repair this damage. Incentive salience theory, used widely to describe reward processing, divides this processing into “liking” (reward-induced hedonic sensory impact) and “wanting” (reward-induced motivation) components. It has been employed to explain pathological changes in reward processing induced by psychiatric disorders. In this review, we summarize the findings of studies of reward processing under chronic pain and examine the effects of chronic pain on “liking” and “wanting.” Evidence indicates that chronic pain compromises the “wanting” component of reward processing; we also discuss the neural mechanisms that may mediate this effect. We hope that this review aids the development of therapies to improve the quality of life of patients with chronic pain.

We investigate the periodic nature of the positive solutions of the fuzzy max-difference equation xn+1=maxAn/xn-m,xn-k,n=0,1,…, where k,m∈{1,2,…}, An is a periodic sequence of fuzzy numbers, and x-d,x-d+1,…,x0 are positive fuzzy numbers with d=m,k. We show that every positive solution of this equation is eventually periodic with period k+1.

Background. The aim of this study was to detect the effect of the laryngeal mask airway (LMA) versus the endotracheal tube (ETT) on postoperative nasal bleedings in and outside the operation room (OR) after intranasal surgery. Methods. 134 patients undergoing elective intranasal surgeries were randomly allocated to receive LMA or ETT during general anesthesia. The incidence, episodes, and severity of nasal bleeding were evaluated in the OR and within the postoperative 24 hours in the ward. Furthermore, medical assistance and severe complications were assessed. Results. The overall incidence of postoperative nasal bleeding throughout the observation period was similar between the two groups. The LMA reduced nasal bleeding in the OR. However, outside the OR, the incidence of the first episode of postoperative nasal bleeding in the LMA group was higher than that in the ETT group (difference: −26.5%; 95% CI: −42.2% to −10.7%; P