Simultaneous bilateral thoracotomy in patients with osteosarcoma and bilateral pulmonary metastases: the experience of the D. Rogachev NMRCPHOI

Osteosarcoma (OS) is the most common primary bone tumor in children and adults. In 15–20% of patients, distant metastases are detected at the time of diagnosis of OS. In more than 80% of cases, metastases are located in the lungs and are the most common disease-related cause of death in OS patients. OS can only be cured if complete surgical remission (CSR) in the lungs is achieved through surgery involving palpation, identification and resection of all detected metastases. Among thoracic surgeons, it is common practice to perform wedge resection of the affected lung parenchyma as it spares more healthy lung tissue. Lobectomy or pneumonectomy can be carried out if either is indicated in the patient. There is, however, no consensus on the best surgical approach for metastasectomy. Our study includes 24 patients who underwent simultaneous bilateral thoracotomy at the Department of Oncology and Pediatric Surgery of the D. Rogachev NMRCPHOI in the period from February 2018 to May 2021. The study was approved by the Independent Ethics Committee and the Scientific Council of the D. Rogachev NMRCPHOI. Eighteen patients underwent primary surgery as part of combination protocol treatment, and six patients were surgically treated for relapse. In 66.7% of the patients treated with upfront surgery, the number of lesions was underestimated, as evident from computed tomography images and intraoperative findings. Post-treatment necrosis grade IV was detected only in 3 patients, in 21.1% of the resected metastases. The median time from bilateral thoracotomy to systemic anti-cancer therapy reinitiation was 12 days. Two patients experienced progression of metastatic disease in the lungs during and immediately the protocol treatment. At the last follow-up, 3 patients were alive with evidence of disease, and 2 patients had died of OS progression. A total of 33.3% of the patients who had had primary surgery developed metastatic (n = 6) and local (n = 1) relapses.

Local Defence System in Healthy Lungs

Background: The respiratory system is one of the main entrance gates for infection. The aim of this work was to compare the appearance of specific mucosal pro-inflammatory and common anti-microbial defence factors in healthy lung tissue, from an ontogenetic point of view. Materials and methods: Healthy lung tissues were collected from 15 patients (three females and 12 males) in the age range from 18 to 86. Immunohistochemistry to human β defensin 2 (HBD-2), human β defensin 3 (HBD-3), human β defensin 4 (HBD-4), cathelicidine (LL-37) and interleukine 17A (IL-17A) were performed. Results: The lung tissue material contained bronchial and lung parenchyma material in which no histological changes, connected with the inflammatory process, were detected. During the study, various statistically significant differences were detected in immunoreactive expression between different factors in all lung tissue structures. Conclusion: All healthy lung structures, but especially the cartilage, alveolar epithelium and the alveolar macrophages, are the main locations for the baseline synthesis of antimicrobial proteins and IL-17A. Cartilage shows high functional plasticity of this structure, including significant antimicrobial activity and participation in local lung protection response. Interrelated changes between antimicrobial proteins in different tissue confirm baseline synergistical cooperation of all these factors in healthy lung host defence.

Determination of initial airtightness after anatomical laser segmentectomy in an ex vivo model

If a pulmonary pathology can be removed by anatomical segmentectomy, the need for lobectomy is obviated. The procedure is considered oncologically equivalent and saves healthy lung tissue. In every segmentectomy, lung parenchyma must be transected in the intersegmental plane. Using an ex vivo model based on porcine lung, three transection techniques (monopolar cutter + suture, stapler, and Nd:YAG laser) are to be compared with respect to their initial airtightness. At an inspiratory ventilation pressure of 25 mbar, all three preparations were airtight. Upon further increase in ventilation pressure up to 40 mbar, the laser group performed best in terms of airtightness. Since thanks to its use of a laser fibre, this technique is particularly suitable for minimally invasive surgery; it should be further evaluated clinically for this indication in the future.

Genetic Deletion of Polo-Like Kinase 2 Induces a Pro-Fibrotic Pulmonary Phenotype

Pulmonary fibrosis is the chronic-progressive replacement of healthy lung tissue by extracellular matrix, leading to the destruction of the alveolar architecture and ultimately death. Due to limited pathophysiological knowledge, causal therapies are still missing and consequently the prognosis is poor. Thus, there is an urgent clinical need for models to derive effective therapies. Polo-like kinase 2 (PLK2) is an emerging regulator of fibroblast function and fibrosis. We found a significant downregulation of PLK2 in four different entities of human pulmonary fibrosis. Therefore, we characterized the pulmonary phenotype of PLK2 knockout (KO) mice. Isolated pulmonary PLK2 KO fibroblasts displayed a pronounced myofibroblast phenotype reflected by increased expression of αSMA, reduced proliferation rates and enhanced ERK1/2 and SMAD2/3 phosphorylation. In PLK2 KO, the expression of the fibrotic cytokines osteopontin and IL18 was elevated compared to controls. Histological analysis of PLK2 KO lungs revealed early stage remodeling in terms of alveolar wall thickening, increased alveolar collagen deposition and myofibroblast foci. Our results prompt further investigation of PLK2 function in pulmonary fibrosis and suggest that the PLK2 KO model displays a genetic predisposition towards pulmonary fibrosis, which could be leveraged in future research on this topic.

COVID-19 lung CT image segmentation using deep learning methods: U-Net versus SegNet

Background Currently, there is an urgent need for efficient tools to assess the diagnosis of COVID-19 patients. In this paper, we present feasible solutions for detecting and labeling infected tissues on CT lung images of such patients. Two structurally-different deep learning techniques, and , are investigated for semantically segmenting infected tissue regions in CT lung images. Methods We propose to use two known deep learning networks, and , for image tissue classification. is characterized as a scene segmentation network and as a medical segmentation tool. Both networks were exploited as binary segmentors to discriminate between infected and healthy lung tissue, also as multi-class segmentors to learn the infection type on the lung. Each network is trained using seventy-two data images, validated on ten images, and tested against the left eighteen images. Several statistical scores are calculated for the results and tabulated accordingly. Results The results show the superior ability of in classifying infected/non-infected tissues compared to the other methods (with 0.95 mean accuracy), while the shows better results as a multi-class segmentor (with 0.91 mean accuracy). Conclusion Semantically segmenting CT scan images of COVID-19 patients is a crucial goal because it would not only assist in disease diagnosis, also help in quantifying the severity of the illness, and hence, prioritize the population treatment accordingly. We propose computer-based techniques that prove to be reliable as detectors for infected tissue in lung CT scans. The availability of such a method in today’s pandemic would help automate, prioritize, fasten, and broaden the treatment of COVID-19 patients globally.

Confocal laser microscopy without fluorescent dye in minimal-invasive thoracic surgery: an ex-vivo pilot study in lung cancer

Cancer will be the leading cause of death in a few decades. In line with minimal invasive lung cancer surgery, surgeons loose most of their tactile tissue information and need an additional tool of intraoperative tissue navigation during surgery. Confocal laser microscopy is a well-established method of tissue investigation. In this ex-vivo pilot study, we evaluated an endoscopic confocal laser microscope (eCLM) that does not need any fluorescent dye as a diagnostic tool in non-malignant and malignant pulmonary tissue and distal stapler resection margins, respectively. In seven cases, an eCLM was used for examining pulmonary tissue ex-vivo. Images of non-malignant and non-small cell lung cancer tissue and distal stapler resection margins were characterized in terms of specific signal-patterns. No fluorescent dye was used. Correlations to findings in conventional histology were systematically recorded and described. Healthy lung tissue showed hyperreflectoric alveolar walls with dark alveolar spaces. Hyperreflective nets indicated the tumor stroma; whereas the hyperreflective areas indicated the tumor cell clusters. Compared to adenocarcinoma tissue, tissue from squamous cell carcinoma showed more distinctive hyperreflective stroma nets. eCLM characteristics seen in non-malignant and malignant tissue were also visible in distal stapler resection margins and so therefore it was feasible to distinguish between healthy lung tissue and lung cancer. This pilot study shows that the assessment of pulmonary tissue with this eCLM for minimally invasive surgical approach without any fluorescent dye is feasible. It enables to differentiate between benign and malignant tissue in pulmonary specimen by easy to evaluate and reproducible parameters.

Covid-19 disease: CT chest volume assessment using nonspecific software “Lung Density” present on “Extended Brilliance Workspace”

The current global COVID-19 pandemic is related to an acute respiratory disease caused by a new coronavirus (SARS-CoV-2), highly contagious and whose evolution is still poorly understood. The high-resolution Computed Tomography (HRCT) is the most accurate technique for identifying pathogenetic finding of interstitial pneumonia. Standardized HRCT examination in COVID patients binds quantitative evaluation of healthy lung tissue performed in post-processing. In this study we present a valid tool for the Radiologist the diagnosis of Covid-19 diagnosis, an essential support in the evaluation of emergency symptomatic patients with negative NAAT, or asymptomatic patients with negative NAAT who have come into contact with positive one, in fact asymptomatic patients can also have lung lesions on CT imaging.

COVID-19 Lung CT Image Segmentation Using Deep Learning Methods: UNET Vs. SegNET

Background: Currently, there is an urgent need for efficient tools to assess the diagnosis of COVID-19 patients. In this paper, we present feasible solutions for detecting and labeling infected tissues on CT lung images of such patients. Two structurally-different deep learning techniques, SegNet and UNET, are investigated for semantically segmenting infected tissue regions in CT lung images.Methods: We propose to use two known deep learning networks, SegNet and UNET, for image tissue classification. SegNet is characterized as a scene segmentation network and UNET as a medical segmentation tool. Both networks were exploited as binary segmentors to discriminate between infected and healthy lung tissue, also as multi-class segmentors to learn the infection type on the lung. Each network is trained using seventy-two data images, validated on ten images, and tested against the left eighteen images. Severalstatistical scores are calculated for the results and tabulated accordingly.Results: The results show the superior ability of SegNet in classifying infected/non-infected tissues compared to the other methods (with 0.95 mean accuracy), while the UNET shows better results as a multi-class segmentor (with 0.91 mean accuracy).Conclusion: Semantically segmenting CT scan images of COVID-19 patients is a crucial goal because it would not only assist in disease diagnosis, also help in quantifying the severity of the illness, and hence, prioritize the population treatment accordingly. We propose computer-based techniques that prove to be reliable as detectors for infected tissue in lung CT scans. The availability of such a method in today’s pandemic wouldhelp automate, prioritize, fasten, and broaden the treatment of COVID-19 patients globally.

Lipids of lung and lung fat emboli of the toothed whales (Odontoceti)

Lipids are biomolecules present in all living organisms that, apart from their physiological functions, can be involved in different pathologies. One of these pathologies is fat embolism, which has been described histologically in the lung of cetaceans in association with ship strikes and with gas and fat embolic syndrome. To assess pathological lung lipid composition, previous knowledge of healthy lung tissue lipid composition is essential; however, these studies are extremely scarce in cetaceans. In the present study we aimed first, to characterize the lipids ordinarily present in the lung tissue of seven cetacean species; and second, to better understand the etiopathogenesis of fat embolism by comparing the lipid composition of lungs positive for fat emboli, and those negative for emboli in Physeter macrocephalus and Ziphius cavirostris (two species in which fat emboli have been described). Results showed that lipid content and lipid classes did not differ among species or diving profiles. In contrast, fatty acid composition was significantly different between species, with C16:0 and C18:1ω9 explaining most of the differences. This baseline knowledge of healthy lung tissue lipid composition will be extremely useful in future studies assessing lung pathologies involving lipids. Concerning fat embolism, non-significant differences could be established between lipid content, lipid classes, and fatty acid composition. However, an unidentified peak was only found in the chromatogram for the two struck whales and merits further investigation.

COVID-19 Lung CT Image Segmentation Using Deep Learning Methods: UNET Vs. SegNET

Background: Currently, there is an urgent need for efficient tools to assess in the diagnosis of COVID-19 patients. In this paper, we present feasible solutions for detecting and labeling infected tissues on CT lung images of such patients. Two structurally-different deep learning techniques, SegNet and UNET, are investigated for semantically segmenting infected tissue regions in CT lung images. Methods: We propose to use two known deep learning networks, SegNet and UNET, for image tissue classification. SegNet is characterized as scene segmentation network and UNet as a medical segmentation tool. Both networks were exploited as binary segmentors to discriminate between infected and healthy lung tissue, and as multi-class segmentors to learn the infection type on the lung. Each network is trained using 72 data images, validated on 10 images and tested against the left 18 images. Several statistical scores are calculated for the results and tabulated accordingly. Results: The results show the superior ability of SegNet in classifying infected/non-infected tissues compared to the other methods (with 0.95 mean accuracy), while the UNET shows better results as a multi-class segmentor (with 0.91 mean accuracy). Conclusion: Semantically segmenting CT scan images of COVID-19 patients is a crucial goal because it would not only assist in disease diagnosis , but also help in quantifying the severity of the disease ,and hence, prioritize the population treatment accordingly. We propose computer-based techniques that prove to be reliable as detectors for infected tissue in lung CT scans. The availability of such a method in today’s pandemic would help automate, prioritize, fasten, and broaden the treatment of COVID-19 patients globally.