Idiopathic subglottic stenosis (iSGS) is a challenging and recurrent disease affecting Caucasian females with an unknown etiology and multiple potential contributing factors.1 The most common method of management of SGS includes endoscopic balloon or rigid dilation, often with carbon dioxide (CO2) laser radial incisions prior to dilation.1 Balloon dilation does carry some risk including tracheal laceration, cricoid fracture, and dilation of healthy trachea leading to potential disruption of normal tracheal mucosal tissue.2 First described in the literature in 2013, CO2 laser wedge excisions are an alternative to traditional balloon dilation. This method involves excision of large wedges of stenotic tissue with preservation of small bridges to allow for re-mucosalization and prevention of circumferential scarring.3 Our institution has employed this technique for nearly 2 decades with promising results. In a retrospective study of iSGS, this method was found to show increased time interval between surgeries compared to standard laser and balloon dilation.1 Recent prospective data has corroborated these findings.4 An in-depth description with surgical photographs regarding this method is lacking and therefore, a more clear explanation with illustrations will be presented. In this procedure, the patient is placed in a standard position for direct laryngoscopy with Aquaplast splints molded around the upper and lower dentition. Laryngeal exposure is typically achieved with a Kleinsasser-C or Dedo laryngoscope. Depending upon surgeon preference, supraglottic jet ventilation through the side port of the laryngoscope or intermittent apnea (with intermittent endotracheal tube placement) is employed. Apneic oxygenation can be helpful in patients at the initiation of the procedure with high flow nasal cannula oxygenation, typically at 60 liters/min. This allows for extra time with sustained high oxygen saturation during laryngeal exposure and the portions of the procedure prior to laser use. Often 1–2 mm of the distal laryngoscope tip is extended through the glottis to carefully spread the posterior glottic aperture allowing for better visualization of the subglottis. Changing the degree of suspension or placing a towel under the head in a “sniffing position” can help with visualization of the posterior and anterior aspects of the stenosis (Fig. 1). A zero-degree Hopkins rod telescope is used for photodocumentation and to measure the length and width of the stenosis as well as proximity to the true vocal folds. The operating microscope with micromanipulator is employed for injections of steroid and wedge excisions. Triamcinolone acetonide suspension (40 mg/cc) is injected into quadrants of the stenosis, often using 1–2 cc in total (Fig. 2). The steroid injection can be done before or after the laser wedge excisions. In our practice, we usually do this in a deeper plane before lasering to avoid spillage of the steroid through the wedge excision sites. If it is performed after laser excisions, then a deeper level near the tracheal and cricoid cartilages is injected both at the islands and troughs. CO2 laser (Lumenis Ultrapulse, in our practice) is used for endoscopic wedge excisions with settings between 2–6 Watts continuous Ultrapulse mode. Palpation of the stenosis is performed with a 90-degree angled probe to ascertain the circumferential depth of the subglottic scar. This helps the surgeon create a three-dimensional understanding of the stenosis and plan where wedge excisions will be made. The more challenging posterior section is often excised first and the easier-to-visualize anterior and lateral portions are excised last. Posterior visualization is best achieved in a more desuspended position aiming towards the posterior cricoid. Manual manipulation of the anterior neck by pulling the skin overlying the cricoid and upper trachea anteriorly while pushing the laryngotracheal complex laterally helps visualize the posterior and lateral portions. Without moving the micromanipulator, manual manipulation of the neck moves the larynx while the other hand holds a 7 French suction that helps control bleeding and provides retraction of tissues. The excision of scar in the subglottis is individualized, with tissue excision directed towards maximizing airway diameter and preservation of the conus elasticus and perichondrium. If the subglottis is stenotic equally around its circumference, then three or four wedges are excised with the CO2 laser leaving narrow bridges in between each excised point (Figs. 3 and 4). Areas that have minimal scar are not lasered, and in these cases fewer bridges are left behind. In a typical stenosis, scar located in the posterior and lateral portion of the airway is addressed first recalling the normal adult airway contour, which is widest in the posterior-lateral confines of the cricoid and upper trachea. Often, the initial incisions at the 4 and 8 o'clock regions are extended distally paralleling where the tracheal rings meet the posterior trachealis mucosa and muscle (Fig. 3A). Scar is excised posterolaterally and then followed inferiorly to the level of the first or second tracheal ring depending on the length of the stenosis (Fig. 3B,C). A narrow island remains at 6 o'clock. A complete posterior excision from 4 o'clock to 8 o'clock can also be excised if the scar is deepest posteriorly but then islands should be preserved at 4 and 8 o'clock. When creating the islands, the base is narrowed significantly to create a minimal stalk attached to the trachea (Fig. 3C). At times, reduction of the island height with CO2 laser can be performed and does not appear to negatively affect healing postoperatively. 0.4 mg/cc mitomycin C is then topically placed with pledgets twice in the subglottis for 2–2.5-minute intervals followed by a 30–60-cc saline rinse with a red rubber catheter, suctioning in the distal trachea as rinsing. Theoretically, the saline rinse allows for less risk of crusting after using mitomycin C, although this is not proven. This is performed with apneic technique instead of jet ventilation. Postoperatively, many of these patients are placed on a long-term medical regimen of one double strength sulfamethoxazole trimethoprim tablet daily, fluticasone 110 mcg inhaler two sprays twice daily, and omeprazole if found to be positive for GERD on pH/impedence testing. See Supplement video S1 for narrated video of the entire procedure. This technique has been utilized for nearly 20 years in our institution. Prospective results from Gelbard et al. demonstrate a significantly lower recurrence rate using endoscopic wedge resection with adjuvant medical therapy (ERMT) compared to endoscopic dilation (ED) (ERMT, 15/121 [12.4%]; and ED, 169/603 [28.0%]). ERMT was associated with temporary tongue paresthesia and rare dental injury.4 CO2 laser endoscopic wedge excisions can be a highly effective method of treating idiopathic subglottic stenosis with decreased recurrence rates compared to traditional endoscopic balloon dilation and improved voice outcomes compared to cricotracheal resection (CTR). Wedge excisions remove disease, unlike ED which moves and compresses tissues. Additionally, ED has been found to have worse voice outcomes on Voice Handicap Index-10 (VHI-10) scores compared with those who underwent ERMT (ED, 10.0 vs. ERMT, 3.5; 95% CI, 0–6.0).4 This difference in voice outcomes is not well understood, but the authors have seen multiple dilated patients who have extension of the subglottic disease to include the glottis. It is not yet known whether this was the natural progression of disease or an iatrogenic cause due to aggressive dilation of both subglottis and glottis. Cricotracheal resection (CTR) showed decreased overall recurrence (1/86, 1.2%) but less superior (360-day) voice outcomes with increased complications including tracheostomy, unplanned return to the operating room, permanent vocal fold paralysis, and anastomotic leak. VHI-10 scores for CTR showed a statistically significant decline compared with scores for patients who underwent ERMT (CTR, 13.0 vs. ERMT, 3.5; 95%CI, 1.0–10.0).4 Historically, circumferential scarring and tracheal perforation was thought to be a risk with aggressive excision in the subglottis, however, prospective evaluation of this technique in fact shows the opposite.4 Small bridges of tissue between the wedge excisions likely are important to prevent this scarring. Eight- to 10-week postoperative evaluation often reveals a near complete dissolution of the island bridges (Figs. 5 and 6). Risk of perforation from laser excision is very low, with no evidence at 3-year follow-up.4 More specifically, posterior perforation has not been seen in our experience likely due to anterior laryngeal suspension allowing for the attached trachealis and esophagus to fall posteriorly from the cricoid inferiorly. The authors recognize several limitations with the EMRT technique. Operative times are longer compared to ED secondary to a learning curve and the greater amount of tissue that is resected. Operative times decrease as the surgeon becomes more comfortable with the procedure. Significant manual manipulation is required to appropriately resect the posterior aspect of the stenosis under microscopic visualization which can pose an ergonomic challenge for some surgeons. Poor endoscopic visualization in certain patients makes this approach more difficult than endoscopic dilation. Patients who may not tolerate longer periods of general anesthesia may not be good candidates for this procedure. Post operatively, airway secretion management can be difficult in some patients. Hypertonic saline nebulized as needed typically resolves the thick mucus and is often necessary for 1 or 2 weeks after surgery. In conclusion, endoscopic CO2 laser wedge resections in combination with maximum medical therapy allow for decrease in recurrence rates, improved voice outcomes and minimal complications in patients with idiopathic subglottic stenosis. Video S1 Endoscopic wedge excisions with CO2 laser for Subglottic Stenosis Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.