Datasets:

---

bluelightai-dev

/

common-corpus-sample-open-science

like 0

Follow

bluelightai-dev 7

ERROR: type should be string, got "https://escholarship.org/uc/item/58g4w73w Copyright Information\nThis work is made available under the terms of a Creative Commons Attribution License, \navailalbe at https://creativecommons.org/licenses/by/4.0/ UC Irvine\nUC Irvine Previously Published Works\nTitle\nDiagnosis of subglottic stenosis in a rabbit model using long-range optical coherence \ntomography.\nPermalink\nhttps://escholarship.org/uc/item/58g4w73w\nJournal\nThe Laryngoscope, 127(1)\nISSN\n0023-852X\nAuthors\nAjose-Popoola, Olubunmi\nSu, Erica\nHamamoto, Ashley\net al.\nPublication Date\n2017\nDOI\n10.1002/lary.26241\nCopyright Information\nThis work is made available under the terms of a Creative Commons Attribution License,\navailalbe at https://creativecommons.org/licenses/by/4.0/\n \nPeer reviewed UC Irvine\nUC Irvine Previously Published Works\nTitle\nDiagnosis of subglottic stenosis in a rabbit model using long-range optical coherence \ntomography. Permalink\nhttps://escholarship.org/uc/item/58g4w73w\nJournal\nThe Laryngoscope, 127(1)\nISSN\n0023-852X\nAuthors\nAjose-Popoola, Olubunmi\nSu, Erica\nHamamoto, Ashley\net al. Publication Date\n2017\nDOI\n10.1002/lary.26241\nCopyright Information\nThis work is made available under the terms of a Creative Commons Attribution License,\navailalbe at https://creativecommons.org/licenses/by/4.0/\n \nPeer reviewed Title Title\nDiagnosis of subglottic stenosis in a rabbit model using long-range optical coherence \ntomography. Permalink Title\nDiagnosis of subglottic stenosis in a rabbit model using long-range optical coherence \ntomography. UC Irvine UC Irvine Previously Published Works UC Irvine Previously Published Works\nTitle\nDiagnosis of subglottic stenosis in a rabbit model using long-range optical coherence \ntomography. Correspondence and request for reprints: Brian J-F. Wong, MD, PhD, Professor and Vice Chairman, Dept. of Otolaryngology-Head & \nNeck Surgery University of California Irvine, 1002 Health Sciences Road, Irvine, CA 92617, bjwong@uci.edu, Phone number: \n949-824-4770, Fax number: 949-824-8413.\n* HHS Public Access\nA h\ni Author Manuscript Published in final edited form as: Published in final edited form as:\nLaryngoscope. 2017 January ; 127(1): 64–69. doi:10.1002/lary.26241. Laryngoscope. 2017 January ; 127(1): 64–69. doi:10.1002/lary.26241. Level of Evidence: NA\nPresented at SPIE Photonics West, San Francisco, CA, USA on February 13, 2016.\nConflict of Interest and Financial Disclosures\nDr. Zhongping Chen has a financial interest in OCT Medical Imaging, Inc., which, however, did not provide support for this work. *These authors contributed equally to the study Powered by the California Digital Library\nUniversity of California eScholarship.org eScholarship.org Abstract Objective—Current imaging modalities lack the necessary resolution to diagnose subglottic \nstenosis. The aim of this study is to use optical coherence tomography (OCT) to evaluate nascent \nsubglottic mucosal injury and characterize mucosal thickness and structural changes using texture \nanalysis in a simulated intubation rabbit model. Author Manuscript Study Design—Prospective animal study in rabbits. Diagnosis of Subglottic Stenosis in a Rabbit Model Using Long-\nRange Optical Coherence Tomography Olubunmi Ajose-Popoola, MD1,*, Erica Su, BS2,*, Ashley Hamamoto, BS2, Alex Wang, BS2, \nJoseph C. Jing, PhD2,3, Tony D. Nguyen, BS2,4, Jason J. Chen, BS2, Kathryn E. Osann, \nPhD4, Zhongping Chen, PhD2,3, Gurpreet S. Ahuja, MD1,5, and Brian J.F. Wong, MD, \nPhD1,2,3 Author Manuscript Presented at SPIE Photonics West, San Francisco, CA, USA on February 13, 2016. Keywords y\nsubglottic stenosis; optical coherence tomography; rabbit model; diagnostic imaging; intubation \ninjury Laryngoscope. Author manuscript; available in PMC 2018 January 01. Study Design—Prospective animal study in rabbits. Methods—Three-centimeter-long sections of endotracheal tubes (ETT) were endoscopically \nplaced in the subglottis and proximal trachea of New Zealand white rabbits (n=10) and secured via \nsuture. OCT imaging and conventional endoscopic video was performed just prior to ETT segment \nplacement (Day 0), immediately after tube removal (Day 7), and one week later (Day 14). OCT \nimages were analyzed for airway wall thickness and textural properties. Results—Endoscopy and histology of intubated rabbits showed a range of normal to edematous \ntissue, which correlated with OCT images. The mean airway mucosal wall thickness measured \nusing OCT was 336.4μm (Day 0), 391.3μm (Day 7), and 420.4μm (Day 14) with significant \ndifferences between Day 0 and Day 14 (p=0.002). Significance was found for correlation and \nhomogeneity texture features across all time points (p<0.05). Author Manuscript Author Manuscript Page 2 Ajose-Popoola et al. Conclusion—OCT is a minimally invasive endoscopic imaging modality capable of monitoring \nprogression of subglottic mucosal injury. This study is the first to evaluate mucosal injury during \nsimulated intubation using serial OCT imaging and texture analysis. OCT and texture analysis \nhave the potential for early detection of subglottic mucosal injury, which could lead to better \nmanagement of the neonatal airway and limit the progression to stenosis. Author Manuscript Introduction Acquired subglottic stenosis (SGS) is most commonly caused by prolonged intubation or \nintubation with an oversized endotracheal tube (ETT). Mucosal injury, as a result of \nischemic necrosis, typically occurs at the subglottis because the cricoid cartilage is a \ncontinuous ring, which does not flex with pressure or allow centrifugal displacement of \ninflammatory edema. Ulceration of the subglottic epithelium may occur within hours of \nintubation and permanent damage is frequently observed within the first 7 days. Although \nthe incidence of neonatal acquired SGS is approximately 0.24–2.0%, the development of \nSGS may have devastating consequences1. Author Manuscript Early diagnosis of SGS is not possible using imaging modalities such as CT and MRI as \nthey lack the necessary imaging resolution. SGS is managed with a range of surgical and \nendoscopic procedures2; however, these procedures may lead to re-stenosis due to induced \ninflammation, requiring additional anti-inflammatory treatments3. Direct laryngoscopy and \nbronchoscopy with dilation is the gold standard for the management of SGS, but it presents \nwith the operative risks of losing a stable, intubated airway and hypoxia. Consequently, the \ntreatment of neonatal SGS remains a diagnostic challenge for otolaryngologists. There have been several SGS models in rabbits using direct trauma to the mucosa (i.e., \nbrushing, caustic chemicals) but very few in evaluating the use of imaging modalities on \nETT injury4. Optical coherence tomography (OCT) is an emerging minimally invasive \noptical imaging technique that provides high resolution (~10 μm) images of living tissue \nbased on different optical properties. OCT has profound impact in diagnosis and \nmanagement of retinal disease in ophthalmology, and has been evaluated intensely in \ncoronary vasculature and GI tract imaging5,6. Traditionally, OCT has focused primarily on \nimaging the upper aerodigestive tract and vocal fold diseases7–9. Our group, along with \nothers, have also focused on the developing pediatric airway as a new application for \nOCT10–12. Author Manuscript Texture analysis is an objective and automated method to obtain quantifiable information \nfrom medical images for the purposes of tissue classification, lesion detection, and \npathologic quantification13. In a digital image, texture is defined as the spatial distribution of \nintensity values, while texture analysis is the evaluation of the relative position and intensity \nof these pixels13. Texture properties derived from gray-level co-occurrence matrices \n(GLCM)—a commonly used statistical approach—have been investigated to characterize Ajose-Popoola et al. Page 3 tissues imaged with OCT14. The main properties successfully used to objectively classify \ntissue include contrast, correlation, homogeneity, and energy14,15. Rabbit Model and Data Acquisition Rabbit Model and Data Acquisition Ethical approval for this study was obtained from the University of California Institutional \nAnimal Care and Use Committee. Ten New Zealand white rabbits (2.4–4.3 kg) were used. At this weight and stage of development, the airway dimensions at the level of the cricoid \n(5.4–5.8 mm) approximate that of neonates (4.5–5.5 mm)17. Author Manuscript The intubation-induced subglottic injury model developed by Kelly et al. was adapted for \nthis study4. A detailed description of the surgical technique and procedure is available in the \nonline supplement. On Day 0, a 3-cm long section of ETT was endoscopically placed just \nbelow the vocal cords (subglottis and proximal trachea) and secured with a suture via a small \ncervical incision. Prior to insertion, the subglottis was sized with an appropriately tight \nfitting ETT (3.5–4.0). OCT and digital video imaging was conducted prior to ETT \nplacement from 3 cm below the subglottis to the level of the glottis. On Day 7, imaging was \nacquired after the segment of ETT was endoscopically removed. On Day 14, the animal was \neuthanized after imaging. The larynx and proximal trachea were dissected free and fixed in \nformaldehyde for 24 hours. The samples were then processed, embedded in paraffin, and \nsequentially sectioned into 6 μm thick slices for histological analysis (hematoxylin-eosin \nstaining). The intubation-induced subglottic injury model developed by Kelly et al. was adapted for \nthis study4. A detailed description of the surgical technique and procedure is available in the \nonline supplement. On Day 0, a 3-cm long section of ETT was endoscopically placed just \nbelow the vocal cords (subglottis and proximal trachea) and secured with a suture via a small \ncervical incision. Prior to insertion, the subglottis was sized with an appropriately tight \nfitting ETT (3.5–4.0). OCT and digital video imaging was conducted prior to ETT Author Manuscript Introduction As OCT studies generate \nhundreds to thousands of images per patient, automated or semi-automated analysis is \nessential, particularly if this technology evolves into a screening measure in the neonatal \nintensive care unit (NICU). Author Manuscript The study objectives are to evaluate OCT as an imaging modality to 1) detect nascent \nsubglottic injury from simulated prolonged intubation in a rabbit model, 2) obtain \nmeasurements of airway wall morphology, and 3) perform automated texture analysis for \ndiagnosis of the condition. Methods\nOCT System Author Manuscript The details of the OCT device and instrumentation used in this study are available in the \nonline supplement and have been previously described16. The flexible OCT probes were \ncustom-built with outer diameters of 1.2 mm and placed within a fluorinated ethylene \npropylene sheath (Zeus Inc., Orangeburg, SC). The probes were then rotated proximally \nwhile being retracted, acquiring images in a retrograde, helical fashion. The system has an \naxial and a lateral resolution of approximately 10 μm and 112 μm, respectively. Laryngoscope. Author manuscript; available in PMC 2018 January 01. Endoscopic and Histologic Analyses Author Manuscript Preintubation endoscopy of all animal subjects was confirmed to be negative for injury in the \nsubglottis prior to the surgical procedure. Control endoscopy of the subglottis was taken to \nallow subsequent comparison (Figure 2A). ETT and suture placement was verified \nendoscopically (Figure 2B). Following extubation, endoscopy of the animal subjects \nuniformly demonstrated edema and inflammation (Figure 2C). Five of the 8 animals \ndemonstrated focal areas of ulceration by Day 14 of the study (Figure 2D). Histologic sections at the level of the cricoid were compared between the intubated group at \nDay 14 and controls from a parallel study19; no animals were euthanized on Day 7. In the \nintubated animals, there was evidence of mucosal thickening in both the epithelium and \nunderlying lamina propria (Figure 3). Regions of the mucosal layer demonstrated signs of \nepithelial metaplasia and abnormal basement membranes with the presence of inflammatory \ncells. In the submucosa, there was indication of increased vascularity and development of \ngranulation tissue. Author Manuscript Morbidity and Mortality Two out of 10 rabbits were excluded from the study: one died as a consequence of a \nwitnessed aspiration during the intubation process, and the other was excluded after one \nweek of intubation as the ETT segment was found to have migrated to the distal trachea. Data Analysis Histologic sections of the subglottis were examined using light microscopy (Microphot-\nFXA; Nikon) and digitized (PictureFrame; Optronics) for documentation. The mucosal and \nsubmucosal layers of the specimen were visually compared to the control specimen from a Ajose-Popoola et al. Page 4 parallel study for changes in thickness and mucosal lining. The sections were also compared \nto corresponding OCT images. Author Manuscript Morphometric and texture analyses were performed on the subglottic region. The tissue \nsegmentation and micrometry methods have been previously described18. Figure 1A shows a \nrepresentative linear image (Figure 1A) and Figure 1B shows a circular image (analogous to \nCT axial images) of the subglottis with manual segmentation in Figure 1C–D. The statistical \nvalues (i.e., mean, standard deviation, and range) of the thickness were calculated across all \naxial-lines of the segmented region of each animal. The average statistical values of each \nexperimental day is reported in Table E1. Texture analysis was performed on the segmented images. Sixteen texture features (contrast, \ncorrelation, energy, and homogeneity) were derived from GLCM (0°, 45°, 90°, and 135°) per \nimage. The mean values per experimental day and their standard error were reported. The \nrepeated measures analysis of variance (rANOVA) was used to measure changes across the 3 \ntime points. Pairwise differences were performed using a one-tailed Student t-test with \nBonferroni correction to adjust for the multiple comparisons. Author Manuscript Laryngoscope. Author manuscript; available in PMC 2018 January 01. OCT Analysis: Soft Tissue Thickness and Texture Analysis Images of Day 0 (control) demonstrated subglottic epithelial layer of low intensity pixels \nwhile the underlying lamina propria presented a uniform layer of higher intensity (Figure \n4A). Darker regions within the submucosal layer indicated the presence of submucosal \nglands or blood vessels, and the first tracheal ring and perichondrium were clearly visible. Laryngoscope. Author manuscript; available in PMC 2018 January 01. Ajose-Popoola et al. Page 5 Images obtained on Day 7, following ETT segment removal, indicated significant pathology \n(Figure 4B). The contrast between the substructures was less pronounced, and the lamina \npropria and submucosa were darker and degraded in appearance. This observation is \nconsistent with granulation tissue and edema10, which were also observed with conventional \ndigital endoscopy. The subglottic airway also demonstrated circumferential narrowing. Images obtained on Day 14 showed marked thickening of the mucosa and submucosa \n(Figure 4C). The contrast between the mucosal substructures was comparable to the control \nand visibly different from images obtained on Day 7; however, the contour of the subglottis \nwas more irregular and focal regions of ulceration were observed. Author Manuscript Author Manuscript Segmentation of the airway cross-section divided the axial tissue structures into 1) the \nairspace, 2) the mucosa and the submucosa, and 3) the cartilage layer. The mean thickness of \nthe mucosa and submucosa layer was 336.4 ± 20.7 μm on Day 0, 391.3 ± 22.9 μm on Day 7, \nand 420.4 ± 15.4 μm on Day 14. There was a significant increase in thickness between Day \n0 and Day 14 of imaging (p=0.002). Between Day 0 and Day 7, the increase was nearly \nsignificant (p=0.053). Figure 5 shows a graphical representation of the mean thickness \nmeasurements for each animal across the 3 imaging days with standard error of the means. Author Manuscript Analysis with rANOVA showed statistically significant increases for correlation and \nhomogeneity variables for all directions across the 3 time points (p<0.05). Running pairwise \ncomparisons between the levels, there was significance between Day 0 and Day 14 for \ncorrelation in all directions (p≤0.016). Changes between Day 0 and Day 7 did not reach \nstatistical significance (p≥0.06 for all directions). For contrast, homogeneity, and energy \nparameters, there were noticeable trends in all directions across the 3 time points. Contrast \nvalues decreased between Day 0 and Day 7 and returned to values more similar to the \ncontrol on Day 14. OCT Analysis: Soft Tissue Thickness and Texture Analysis Energy and homogeneity values increased between Day 0 and Day 7 and \nreturned to more similar values to the control on Day 14. Details of the statistical analysis \nare found in Table E1. Author Manuscript Laryngoscope. Author manuscript; available in PMC 2018 January 01. Discussion In our previous study in critically ill intubated neonates in the NICU, OCT measurement of \nsoft tissue changes in the subglottic mucosa was reported20. However, we were not able to \ncorrelate endoscopic images or histopathology, as the number of subjects progressing to \noperative endoscopy was small, and full thickness (including cartilage) surgical specimens \nare rarely available in these subjects20. In this study, the animal model was used to simulate \nthe clinical scenario (intubation) to provide insight into the relationship between OCT, \nhistologic, and endoscopic observations. Early injury due to prolonged intubation cannot be \ndiagnosed using conventional imaging modalities (CT, MRI, etc.) due to their limited \nimaging resolution. Direct laryngoscopy is the gold standard of the diagnosis of SGS, but is \ngenerally performed under general anesthesia, which could result in complete airway \nobstruction. Alternatively, OCT can be performed daily as a screening measure through the \nendotracheal tube and would not require extubation or endoscopy. With its high resolution, \nOCT may detect early signs of SGS, providing a means to better manage the ETT, prevent \nprogression to high-grade SGS, and even predict patients with high risk for extubation \nfailure. Author Manuscript Ajose-Popoola et al. Page 6 Page 6 This study illustrates the potential for OCT to detect mucosal injury and thickening in the \nsubglottis. Soft tissue layer thickness measurements and texture analysis demonstrated \nstatistically significant increases in mucosal and submucosal thickness and changes in \ntexture parameters over time, respectively. The model used here best replicates the clinical \ncircumstances, where the ETT remains in the airway for a prolonged time period. Texture \nanalysis has been used in many settings particularly in remote sensing and machine vision \napplications21. Here, texture analysis is demonstrated as an automated means to analyze \ntissue. Author Manuscript\nA Author Manuscript Previous SGS models with arduous methods (i.e., electrocautery, chemical irritation, curette,\nlaser, nylon brush) produced high-grade stenoses associated with high mortality rates and \ndid not mimic clinical neonatal SGS19,22–25. Actual intubation models have been used, \nhowever only acute changes can be studied due to limitations with respect to how long a \nrabbit can remain intubated26. In contrast, the ETT model developed by Kelly et al. has \ndemonstrated to increase lamina propria thickness, mucosal thickness, and goblet cell \ndensity4. The mucosal injuries induced in this study were similar, though no perichondritis \nor chondritis was observed. Laryngoscope. Author manuscript; available in PMC 2018 January 01. Discussion This is possibly due to differences in ETT sizing, as the degree \nof injury depends on the pressure induced by the oversized ETT. Regardless, ulceration, \ngranulation tissue, and mucosal thickening were observed. The superficial injury created \nhere is more similar to what is believed to occur during the early stages of subglottic injury \nin the neonate, which is seldom clinically documented as operative endoscopy is rarely, if \never, performed at this time in disease evolution. Additionally, OCT tracked mucosal \nthickening over 3 time points in the same animal, while other animal studies have relied \nentirely upon histology obtained only at euthanasia. Author Manuscript A limiting factor in this study was not including micrometry of histologic sections at \ndifferent time points as histology is only available on Day 14, the time in which euthanasia \noccurred. Animal subject numbers were kept limited because previous studies have already \ndemonstrated the link between OCT-based measurements of mucosal thickness and \nhistology19. Furthermore, micrometry based upon histologic sections is fraught with artifact \ndue to fixation methods, whereas OCT micrometry is precise to less than 10 μm27. Mucosal thickness was found to increase through one week of intubation and continued to \nincrease after extubation. There was statistical significance between measurements on Day 0 \nand Day 14, near significance between Day 0 and Day 7, and no significance between Day 7 \nand 14. This suggests that most of the mucosal injury occurs within the week of intubation \nwhile the inflammatory process may have continued even after extubation. These findings \ncorrelate with the signs of early SGS, including ischemia and edema caused by prolonged \nmucosal compression, then ulceration, inflammation, and granulation tissue formation \nfollowing persistent injury. After extubation, the healing of subglottic ulcers continues and \nfollows a similar pathway as wound healing of the skin28. Author Manuscript Each OCT scan includes hundreds of images, and if used as a clinical screening measure, \nautomated or semi-automated processes may be required. In this study, labor intensive \nmanual segmentation of distinct tissue regions was performed by one investigator (E.S.). The feasibility of texture analysis using OCT images for the purposes of tissue classification Ajose-Popoola et al. Page 7 Page 7 have been reported, though use has not been widespread14,29. GLCM properties potentially \nallow differentiation of tissue type without manual segmentation14 and of tissue phantoms \ncontaining various sizes and distributions of scattering foci29. Supplementary Material Refer to Web version on PubMed Central for supplementary material. Author Manuscript Conclusion This study is the first to characterize ETT injury to the rabbit airway using OCT and apply \ntexture analysis to detect changes in tissue composition. OCT allows for quantifiable \nmeasurements of airway soft tissue thickening, showing statistically significant changes \nacross different time points. Texture analysis provides a basis to characterize the state of the \npathology relative to the control, and the success herein makes automated classification of \npathology possible. Such an automated approach would be essential should OCT evolve to \nbecome a common part of ETT and airway management in the NICU. Future studies will \nfocus on combining this ETT SGS injury model with pepsin and hydrochloric acid \napplication to simulate acid reflux. Author Manuscript Discussion To our knowledge, our \ninvestigation is the first to apply texture analysis to airway OCT images. Changes in the \ntexture statistics show measurable differences between pathological states of the intubated \nsubglottis. The significant changes in correlation and homogeneity herein demonstrate \npotential for automated diagnosis. In addition, the trends in homogeneity and energy indicate \nan increase in the value after one week of intubation followed by a return to values similar to \nthe control after one week following extubation. Contrast showed a decrease followed by a \nreturn to values similar to the control. Author Manuscript Acknowledgments This study was supported by National Institutes of Health (NIH) award numbers: 1-R01-HL103764-01 and 1-R01-\nHL105215-01 from the National Heart, Lung and Blood Institute (NHLBI), 1-R43-HD071701-01 from the National \nInstitute of Child Health and Human Development (NICHD), and P41EB015890 from the National Institute of \nBiomedical Imaging and Bioengineering (NIBIB). References 1. Walner DL, Loewen MS, Kimura RE. Neonatal Subglottic Stenosis-Incidence and Trends. Laryngoscope. 2001; 111(1):48–51. DOI: 10.1097/00005537-200101000-00009 [PubMed: \n11192899] 2. Wei JL, Bond J. Management and prevention of endotracheal intubation injury in neonates. Curr \nOpin Otolaryngol Head Neck Surg. 2011; 19(6):474–477. DOI: 10.1097/MOO.0b013e32834c7b5c \n[PubMed: 21986802] Author Manuscript 3. Ingrams DR, Ashton P, Shah R, Dhingra J, Shapshay SM. Slow-release 5-fluorouracil and \ntriamcinolone reduces subglottic stenosis in a rabbit model. Ann Otol Rhinol Laryngol. 2000; \n109(4):422–424. [PubMed: 10778898] 3. Ingrams DR, Ashton P, Shah R, Dhingra J, Shapshay SM. Slow-release 5-fluorouracil and \ntriamcinolone reduces subglottic stenosis in a rabbit model. Ann Otol Rhinol Laryngol. 2000; \n109(4):422–424. [PubMed: 10778898] 4. Kelly NA, Murphy M, Giles S, Russell JD. Subglottic injury: A clinically relevant animal model. Laryngoscope. 2012; 122(11):2574–2581. DOI: 10.1002/lary.23515 [PubMed: 22961393] 4. Kelly NA, Murphy M, Giles S, Russell JD. Subglottic injury: A clinically relevant animal model. Laryngoscope. 2012; 122(11):2574–2581. DOI: 10.1002/lary.23515 [PubMed: 22961393] 5. Gerckens U, Buellesfeld L, McNamara E, Grube E. Optical coherence tomography (OCT): Potential \nof a new high-resolution intracoronary imaging technique. Herz. 2003; 28(6):496–500. DOI: \n10.1007/s00059-003-2485-9 [PubMed: 14569390] 5. Gerckens U, Buellesfeld L, McNamara E, Grube E. Optical coherence tomography (OCT): Potential \nof a new high-resolution intracoronary imaging technique. Herz. 2003; 28(6):496–500. DOI: \n10.1007/s00059-003-2485-9 [PubMed: 14569390] Laryngoscope. Author manuscript; available in PMC 2018 January 01. Ajose-Popoola et al. Page 8 6. Costa RA, Skaf M, Melo LAS, et al. Retinal assessment using optical coherence tomography. Prog \nRetin Eye Res. 2006; 25(3):325–353. DOI: 10.1016/j.preteyeres.2006.03.001 [PubMed: 16716639] Author Manuscript 7. Sepehr A, Armstrong WB, Guo S, et al. Optical coherence tomography of the larynx in the awake \npatient. Otolaryngol - Head Neck Surg. 2008; 138(4):425–429. DOI: 10.1016/j.otohns.2007.12.005 \n[PubMed: 18359348] 8. Kobler JB, Chang EW, Zeitels SM, Yun SH. Dynamic imaging of vocal fold oscillation with four-\ndimensional optical coherence tomography. Laryngoscope. 2010; 120(7):1354–1362. DOI: 10.1002/\nlary.20938 [PubMed: 20564724] 9. Ridgway JM, Armstrong WB, Guo S, et al. In vivo optical coherence tomography of the human oral \ncavity and oropharynx. Arch Otolaryngol Head Neck Surg. 2006; 132(10):1074–1081. DOI: \n10.1001/archotol.132.10.1074 [PubMed: 17043254] 10. Ridgway JM, Ahuja G, Guo S, et al. Imaging of the Pediatric Airway Using Optical Coherence \nTomography. 2007; 117doi: 10.1097/MLG.0b013e318145b306 11. Lazarow FB, Ahuja GS, Chin Loy A, et al. Intraoperative long range optical coherence tomography \nas a novel method of imaging the pediatric upper airway before and after adenotonsillectomy. Int J \nPediatr Otorhinolaryngol. 2015; 79(1):63–70. Laryngoscope. Author manuscript; available in PMC 2018 January 01. References DOI: 10.1016/j.ijporl.2014.11.009 [PubMed: \n25479699] Author Manuscript 12. Volgger V, Sharma GK, Jing JC, et al. Long-range Fourier domain optical coherence tomography \nof the pediatric subglottis. Int J Pediatr Otorhinolaryngol. 2015; 79(2):119–126. DOI: 10.1016/\nj.ijporl.2014.11.019 [PubMed: 25532671] 13. Castellano G, Bonilha L, Li LM, Cendes F. Texture analysis of medical images. Clin Radiol. 2004; \n59:1061–1069. DOI: 10.1016/j.crad.2004.07.008 [PubMed: 15556588] 14. Gossage KW, Tkaczyk TS, Rodriguez JJ, Barton JK. Texture analysis of optical coherence \ntomography images: feasibility for tissue classification. J Biomed Opt. 2003; 8(3):570–575. DOI: \n10.1117/1.1577575 [PubMed: 12880366] 15. Lin Y-S, Chu C-C, Lin J-J, et al. Optical coherence tomography: A new strategy to image planarian \nregeneration. Sci Rep. 2014; 4:6316.doi: 10.1038/srep06316 [PubMed: 25204535] 16. Jing J, Zhang J, Loy AC, Wong BJF, Chen Z. High-speed upper-airway imaging using full-range \noptical coherence tomography. J Biomed Opt. 2012; 17(11):110507.doi: 10.1117/1.JBO. 17.11.110507 [PubMed: 23214170] Author Manuscript 17. Loewen MS, Walner DL. Dimensions of rabbit subglottis and trachea. Lab Anim. 2001; 35(3):253–\n256. DOI: 10.1258/0023677011911714 [PubMed: 11459410] 18. Su, E., Sharma, GK., Chen, J., et al. Analysis and digital 3D modeling of long-range fourier-\ndomain optical coherence tomography images of the pediatric subglottis. In: Choi, B.Kollias, \nN.Zeng, H., et al., editors. Proc. SPIE 8926, Photonic Therapeutics and Diagnostics X; 2014. p. 89262J 19. Lin JL, Yau AY, Boyd J, et al. Real-Time Subglottic Stenosis Imaging Using Optical Coherence \nTomography in the Rabbit. JAMA Otolaryngol Neck Surg. 2013; 139(5):502.doi: 10.1001/\njamaoto.2013.2643 20. Sharma GK, Ahuja GS, Wiedmann M, et al. Long Range Optical Coherence Tomography of the \nNeonatal Upper Airway for Early Diagnosis of Intubation-related Subglottic Injury. Am J Respir \nCrit Care Med. Jul.2015 150727130422005. doi: 10.1164/rccm.201501-0053OC 21. du Buf JMH, Kardan M, Spann M. Texture feature performance for image segmentation. Pattern \nRecognit. 1990; 23(3–4):291–309. DOI: 10.1016/0031-3203(90)90017-F Author Manuscript 22. Cincik H, Gungor A, Cakmak A, et al. The effects of mitomycin C and 5-fluorouracil/\ntriamcinolone on fibrosis/scar tissue formation secondary to subglottic trauma (experimental \nstudy). Am J Otolaryngol. 2005; 26(1):45–50. [PubMed: 15635581] 23. Roh JL, Lee YW, Park HT. Subglottic wound healing in a new rabbit model of acquired subglottic \nstenosis. Ann Otol Rhinol Laryngol. 2006; 115(8):611–616. [PubMed: 16944660] 24. Steehler MK, Hesham HN, Wycherly BJ, Burke KM, Malekzadeh S. Induction of tracheal stenosis \nin a rabbit model-endoscopic versus open technique. Laryngoscope. 2011; 121(3):509–514. DOI: \n10.1002/lary.21407 [PubMed: 21344426] Laryngoscope. Author manuscript; available in PMC 2018 January 01. Ajose-Popoola et al. Page 9 25. Laryngoscope. Author manuscript; available in PMC 2018 January 01. Laryngoscope. Author manuscript; available in PMC 2018 January 01. References Chafin JB, Sandulache VC, Dunklebarger JL, et al. Graded carbon dioxide laser-induced subglottic \ninjury in the rabbit model. Arch Otolaryngol Head Neck Surg. 2007; 133(4):358–364. DOI: \n10.1001/archotol.133.4.358 [PubMed: 17438250] Author Manuscript 26. Kumar SP, Ravikumar A, Thanka J. An animal model for laryngotracheal injuries: An experimental \nstudy. Laryngoscope. 2015; 125(1):E23–E27. DOI: 10.1002/lary.24867 [PubMed: 25111795] 27. Kaiser ML, Rubinstein M, Vokes DE, et al. Laryngeal epithelial thickness: A comparison between \noptical coherence tomography and histology. Clin Otolaryngol. 2009; 34(5):460–466. DOI: \n10.1111/j.1749-4486.2009.02005.x [PubMed: 19793279] 28. Eid, Ea. Anesthesia for subglottic stenosis in pediatrics. Saudi J Anaesth. 2009; 3(2):77–82. DOI: \n10.4103/1658-354X.57882 [PubMed: 20532108] 29. Gossage KW, Smith CM, Kanter EM, et al. Texture analysis of speckle in optical coherence \ntomography images of tissue phantoms. Phys Med Biol. 2006; 51(6):1563–1575. DOI: \n10.1088/0031-9155/51/6/014 [PubMed: 16510963] Author Manuscript Author Manuscript Author Manuscript Ajose-Popoola et al. Page 10 Figure 1. Representative OCT image of the rabbit subglottis in linear (A) and circular representation \n(C). Sample segmented images in linear (B) and circular representation (D). Scale bar \nrepresents 2 mm. Definition of abbreviations: TC = tracheal cartilage, LP = lamina propria, \nPS = outer probe sheath. Author Manuscript Author Manuscript Figure 1. Figure 1. Representative OCT image of the rabbit subglottis in linear (A) and circular representation \n(C). Sample segmented images in linear (B) and circular representation (D). Scale bar \nrepresents 2 mm. Definition of abbreviations: TC = tracheal cartilage, LP = lamina propria, \nPS = outer probe sheath. Author Manuscript Author Manuscript Ajose-Popoola et al. Page 11 Figure 2. Endoscopic footage of gross morphology of the subglottis. (A) The control subglottic region \nand (B) subglottic region after tube placement with clear visualization of the suture. (C) \nEdema and inflammation apparent after one week of intubation and (D) focal areas of \nulceration after one week of extubation. Figure 2. Endoscopic footage of gross morphology of the subglottis. (A) The control subglottic region \nand (B) subglottic region after tube placement with clear visualization of the suture. (C) \nEdema and inflammation apparent after one week of intubation and (D) focal areas of \nulceration after one week of extubation. Author Manuscript Author Manuscript\nAu Author Manuscript Figure 2. Endoscopic footage of gross morphology of the subglottis. (A) The control subglottic region \nand (B) subglottic region after tube placement with clear visualization of the suture. Laryngoscope. Author manuscript; available in PMC 2018 January 01. Laryngoscope. Author manuscript; available in PMC 2018 January 01. Laryngoscope. Author manuscript; available in PMC 2018 January 01. References (C) \nEdema and inflammation apparent after one week of intubation and (D) focal areas of \nulceration after one week of extubation. Figure 2. Endoscopic footage of gross morphology of the subglottis. (A) The control subglottic region \nand (B) subglottic region after tube placement with clear visualization of the suture. (C) \nEdema and inflammation apparent after one week of intubation and (D) focal areas of \nulceration after one week of extubation. Author Manuscript Author Manuscript Page 12 Ajose-Popoola et al. Figure 3. Histologic comparison between control and intubated specimen. (A) The normal mucosal \nand submucosal layers of an unintubated animal and (B) thickened mucosal and submucosal \nlayers of an intubated animal sacrificed after one week of extubation. Scale bars for (A) and \n(B) represent 250 μm. (C) Entire section of the normal subglottis at the level of the cricoid \ncartilage. (D) Entire section of the intubated subglottis with evidence of edema and \ngranulation tissue. Scale bars for (C) and (D) represent 2 mm. Author Manuscript Figure 3. Histologic comparison between control and intubated specimen. (A) The normal mucosal \nand submucosal layers of an unintubated animal and (B) thickened mucosal and submucosal \nlayers of an intubated animal sacrificed after one week of extubation. Scale bars for (A) and \n(B) represent 250 μm. (C) Entire section of the normal subglottis at the level of the cricoid \ncartilage. (D) Entire section of the intubated subglottis with evidence of edema and \ngranulation tissue. Scale bars for (C) and (D) represent 2 mm. Author Manuscript Author Manuscript Author Manuscript Page 13 Page 13 Ajose-Popoola et al. Figure 4. OCT images of the subglottis at different time points. (A) Control subglottic region showing \nnormal anatomy. (B) Thickened mucosal layer of the subglottic region with evidence of \nedema (darker regions) after one week of intubation. (C) Thickened mucosal and \nsubmucosal layers after one week of extubation. Scale bar represents 2 mm. Author Manuscript Figure 4. g\nOCT images of the subglottis at different time points. (A) Control subglottic region showing \nnormal anatomy. (B) Thickened mucosal layer of the subglottic region with evidence of \nedema (darker regions) after one week of intubation. (C) Thickened mucosal and \nsubmucosal layers after one week of extubation. Scale bar represents 2 mm. Author Manuscript Author Manuscript Author Manuscript Page 14 Ajose-Popoola et al. Page 14 Figure 5. Graphical representation of the effect of intubation on the mucosal and submucosal \nthickness. Laryngoscope. Author manuscript; available in PMC 2018 January 01. References Mean airway soft tissue thickness measurements of all animals are shown with \nsignificant increase in thickness between Day 0 and Day 14 (*p = 0.002). Author Manuscript Author Manuscript Author Manuscript Figure 5. Graphical representation of the effect of intubation on the mucosal and submucosal \nthickness. Mean airway soft tissue thickness measurements of all animals are shown with \nsignificant increase in thickness between Day 0 and Day 14 (*p = 0.002). Author Manuscript"