Minimally Invasive Murine Laryngoscopy for Close-Up Imaging of Laryngeal Motion During Breathing and Swallowing

The murine laryngoscopy protocol follows an approved Institutional Animal Care and Use Committee (IACUC) protocol and National Institutes of Health (NIH) Guidelines. It was developed for use with over 100 adult C57BL/6J mice and over 50 adult Sprague Dawley rats, approximately equal sexes and 6 weeks-12 months old for both species. Additional protocol development is necessary for adaptation to younger/smaller rodents. Animals were group housed (up to four mice or two rats per cage, based on sex and litter). The standard vivarium conditions included static caging with strict regulation of ambient temperature (20-26 °C), humidity (30%-70%), and standard 12 h light cycle. All animals received fresh enrichment materials (e.g., hut/pipe, dental treats, nestlet) at weekly cage changes. Unlimited access to food and water was provided, except during a short (up to 4-6 h) food restriction prior to anesthesia as described below. Veterinary and research staff monitored the animals every day.

1. Animal anesthesia that does not abolish swallowing

1. Wear appropriate personal protective equipment (e.g., gloves, mask) to minimize allergen exposure while working with rodents.
2. Food-restrict each rodent cage up to 4-6 h before anesthesia to minimize the retention of food in the oral cavity and pharynx, which may interfere with endoscopic visualization and/or result in food aspiration during the procedure.
   NOTE: Food retention in the oral cavity is a normal finding in rodents without dysphagia if they have not been food-restricted.
3. Prepare a "warming station" for the induction/recovery of the animals.
   1. Prewarm a water-circulating heating pad to 37 °C on a benchtop surface.
   2. Select appropriately sized induction/recovery cages for the species being tested. For example, mouse shoebox cages with filter top lids are appropriately sized for the induction/recovery of mice and rats. Use a fresh induction/recovery cage for each animal being tested; use a single cage as both the induction and recovery cage for the same animal.
   3. Line the induction/recovery cage floor with a light layer of absorbent material (e.g., aspen shavings, paper towel, puppy pad) for warmth and absorption of body secretions during anesthesia induction and recovery.
   4. Place the prepared cages (with filter top lids) fully on the heating pad for 30-60 min prior to anesthesia induction.
      NOTE: This microenvironment provides sufficient supplemental heat to promote stable anesthesia metabolism during induction and recovery.
4. Place the animal's home cage halfway on the preheated 37 °C heating pad for approximately 30 min prior to anesthesia induction.
   NOTE: Providing supplemental heat in advance of the procedure may hasten anesthesia induction and prevent accidental overdose due to slowed/delayed anesthesia metabolism from hypothermia.
5. Prepare ketamine-xylazine (KX) anesthesia based on the species and body weight.
   1. For mice: A mixture of 90 mg/kg ketamine and 11 mg/kg xylazine is sufficient for transoral laryngoscopy in adult C57BL/6 background mice of either sex. Adjust the doses for other mouse strains and ages.
   2. For rats: A mixture of 60 mg/kg ketamine and 6 mg/kg xylazine is sufficient for transoral laryngoscopy in adult Sprague Dawley rats of either sex. Adjust the doses for other rat strains and ages.
6. Prewarm the syringe-filled anesthetic agents on the 37 °C warming station to prevent heat loss in the animals that occurs when injecting cold fluids.
7. Inject the rodent with the calculated KX dose using an appropriately sized syringe (e.g., 1 mL) and needle (e.g., 26 G½).
   1. For mice: Administer a single subcutaneous (SC) injection.
      NOTE: In our experience, SC injections in mice reduce/abolish anesthesia-related mortality compared to intraperitoneal (IP) injections.
   2. For rats: Administer a single SC or IP injection. If preferred, sedate rats with isoflurane (ISO) (3-5%) in an induction chamber immediately prior to KX injection.
      NOTE: Spontaneous body movement may resume for a brief period (typically <1 min) until the KX takes effect.
8. Administer glycopyrrolate (anticholinergic agent) immediately after the KX injection to reduce excess salivary secretions that may hinder visualization of the larynx during transoral endoscopy and/or mechanically obstruct the upper airway during anesthesia recovery.
   NOTE: The dosage and delivery route are identical for mice and rats (0.01-0.02 mg/kg SC), and the effect is nearly immediate and lasts several hours.
9. After glycopyrrolate injection, place the anesthetized rodent in the preheated induction cage on the warming station and cover the cage with a surgical drape to provide a darkened environment that minimizes visual stimulation for 10 min, undisturbed.
   1. Terminate the procedure if the rodent remains ambulatory 10 min after the initial KX dose.
      NOTE: Attempts to provide additional anesthesia (either KX or ketamine maintenance) will likely be futile.
10. After 10 min, administer a maintenance dose of ketamine (¼ of the initial dose if hindlimb reflexes are diminished or ½ the initial dose if hindlimb reflexes are brisk; SC for mice and SC/IP for rats) to maintain anesthesia.
11. Apply ophthalmic ointment to both eyes to prevent corneal drying and associated trauma during the laryngoscopy procedure.
12. Transfer the anesthetized rodent to a custom endoscopy platform to begin the laryngoscopy procedure.
    NOTE: We designed our endoscopy platform (Figure 1) to have multiple functionalities for use with a variety of rodent surgical and electrophysiology approaches that do not necessarily require endoscopy. As such, it is overbuilt for purely endoscopic use. Where relevant, we will highlight features/components that are essential to this laryngoscopy protocol.
13. From this point forward, check hindlimb reflexes every 15-20 min throughout the entire procedure and provide additional ketamine maintenance doses as needed, spaced at least 20 min apart. As this is a relatively short procedure (typically <45 min under anesthesia), additional ketamine is rarely required after the initial maintenance dose.

2. Transoral passage of the endoscope to visualize the larynx

1. Prior to anesthetizing the animal, prepare an appropriately sized endoscope with video-recording capability.
   ​NOTE: We routinely use a zero-degree Otoscope with a 1.9 mm shaft diameter and 10 cm shaft length with a custom metal sheath (Figure 2), which is the representative endoscope utilized throughout this protocol.
   1. Connect the endoscope to a light source and endoscopy recording system for real-time viewing and video recording at a minimum of 30 frames per second (fps).
   2. Focus and white balance the camera for optimal image quality.
   3. Attach the endoscope to a micromanipulator.
      NOTE: For laryngeal motion tracking, we secure the endoscope to a micromanipulator on the endoscopy platform to permit precise endoscope control for stable video capture.
2. Secure the rodent in dorsal recumbency on a heated platform. Stabilize and immobilize the head by securing it with ear bars.
   1. Ensure the rodent's head can freely rotate up/down (but not side-to-side) in the ear bars without slipping out. This degree of freedom facilitates transoral insertion and advancement of the endoscope to reach the larynx.
3. If synchronous electrophysiological recording of breathing, swallowing, and swallow-breathing coordination is desired during endoscopy, proceed with applying appropriate sensors for this purpose (Figure 3).
   1. Secure a respiratory sensor to the abdomen at midline, immediately caudal to the xyphoid process, using surgical tape.
   2. Shave and clean/disinfect the skin with an alcohol wipe before inserting the needle electrodes to prevent infection.
   3. Use a 22 G needle to pierce a small opening through the skin before inserting the needle electrode to prevent electromyography (EMG) needle damage.
   4. Insert a sterile concentric EMG needle electrode (e.g., 25 mm x 0.3 mm/30 G) through the submental skin at midline into the tongue base (e.g., genioglossus or geniohyoid muscle, depending on needle insertion depth).
   5. Insert a ground electrode (e.g., 27 G stainless steel) subcutaneously at the hip (either side).
   6. Connect the respiratory sensor and EMG needle electrodes to an electrophysiology recording system (e.g., bioamplifier and data acquisition system with synchronous video capture) and verify clean electrophysiology signals in both channels before proceeding.
   7. Wrap the electrode connection sites with aluminum foil to shield against electrical noise and improve the signal-to-noise ratio in the corresponding electrophysiological recordings.
   8. Adjust the respiratory sensor location and EMG needle electrode depth as needed to obtain clean electrophysiology signals in both channels. To follow this protocol, use a 1k sampling rate for respiration and a 20k sampling rate and band pass filter (e.g., 150-3,000 Hz) for EMG.
4. Cover the rodent's torso (and respiratory sensor) with a transparent blanket to facilitate thermoregulation while permitting visualization of abdominal motion during respiration. Leave the hindlimbs and lower abdomen freely exposed for access during reflex checks and ketamine maintenance redosing. Ensure the blanket does not restrict abdominal movement during breathing.
5. Proceed with transoral endoscopy (Figure 4).
   1. Open the rodent's mouth by inserting a tapered, cotton-tipped applicator behind the central incisors, perpendicular to the jaw. Rotate the cotton swab on the dorsal surface of the tongue to slightly protrude it from the mouth.
6. Using a light finger grip, gently pull the tongue slightly out of the mouth to one side of the central incisors while inserting the endoscope tip into the oral cavity (Figure 4A,B).
   1. Turn on the light source after inserting the endoscope tip into the mouth to avoid potentially harming the rodent's eyes.
   2. Insert the endoscope lateral to the incisors on the same side as the retracted tongue. The central incisors prevent the endoscope from being inserted at midline, thus necessitating this lateral insertion approach.
7. Start the endoscopy (and electrophysiology) recording systems. Record continuously throughout the entire procedure to ensure sufficient data for post hoc analysis or record at select times, depending on the study's needs.
8. Carefully advance the endoscope to visualize the oropharynx, using caution not to scrape against the hard palate or apply excess pressure on the tongue that may cause injury.
   1. Remove any visible food particles and/or excess salivary secretions using an appropriately sized swab (e.g., a 1.5 mm micro-brush) to minimize aspiration risk as the procedure advances.
   2. Continue advancing/adjusting the endoscope position until the hypopharynx is centered within the field of view on the monitor and key anatomical structures are identifiable (Figure 4C). At this point, all structures should appear anatomically aligned/symmetric within the camera field of view; otherwise, reposition the endoscope as needed.
   3. Watch for evoked jaw/tongue movement during endoscope advancement. If absent, proceed without further ketamine redosing. If present, administer a second ketamine maintenance dose (¼ to ½ of the initial dose of ketamine) and wait approximately 5 min to take effect before proceeding. Re-dose only if it has been at least 20 min since the previous injection to avoid oversedation and abolishment of swallowing.
   4. Examine the rodent's tongue every 5 min throughout the procedure for darkened discoloration, which is indicative of ischemia. To avoid this, reposition the endoscope as needed.
9. Apply light pressure to the velum with a micro-probe (e.g., a metal spatula) inserted alongside the endoscope to uncouple the soft palate and epiglottis to visualize the larynx from a distance (Figure 4D). Avoid using the endoscope tip for decoupling, as the applied pressure may cause soft tissue damage or permanently bend/damage the endoscope shaft.
   NOTE: Unlike humans, the murine larynx is not directly visible from a transoral perspective. Instead, the epiglottis is mechanically entrapped beneath a mucosal membrane overlying the velum, which results in the formation of a cul-de-sac hypopharyngeal space. Applying light pressure on the velum releases the epiglottis from the velar membrane to give a partial view of the larynx.
10. Observe for evoked swallows during velum/epiglottis decoupling.
    1. Identify swallows as abrupt, brief posterior tongue displacements toward the hard palate. This movement typically occurs in synchrony with brief mandibular movement/depression, thus providing a surrogate for the identification of swallowing when the posterior tongue is not readily visible in the endoscope field of view.
    2. Also identify swallows via tongue EMG bursting activity in conjunction with brief apneic episodes in the electrophysiology recording, both occurring in synchrony with glottic closure events in the endoscopy video.
       1. In the case of rapid repetitive swallowing indicative of insufficient anesthesia (i.e., too light), re-dose and wait ~5-10 min before proceeding. Wait at least 20 min after the previous ketamine maintenance injection to avoid oversedation and abolishment of swallowing.
       2. Consider the anesthetic depth to be optimal when only a few swallows are evoked during velar-epiglottic decoupling.
       3. If swallowing is abolished, anesthesia is too deep for assessing laryngeal airway protection. In this case, wait 5-10 min for ketamine metabolism before proceeding with close-up visualization of the larynx.

3. Close-up, unobstructed video recording of laryngeal motion during breathing and evoked swallowing

NOTE: Synchronous electrophysiological recording of breathing, swallowing, and swallow-breathing coordination is also an option.

1. Slowly advance the endoscope between the velum and epiglottis while maintaining the larynx in the center of the field of view (Figure 5A-C).
   NOTE: The endoscope tip will readily pass through the velar-epiglottic opening without force. Otherwise, abort the procedure to avoid potential harm to the animal. It is possible to visualize the larynx from a distance, with the endoscope tip in the hypopharynx. However, this approach typically requires manual retraction of the epiglottis, velum, and/or tongue for enhanced visualization of the larynx. However, portions of the larynx typically remain obscured from view, and the retraction devices can restrict laryngeal motion, which may be mistaken as dysfunction.
2. Continue advancing the endoscope to obtain an unobstructed close-up view of the entire ventral-dorsal and lateral dimensions of the larynx in a single field of view (Figure 5C).
   NOTE: The ventral commissure may be obstructed by the epiglottis, particularly in younger/smaller mice. In these cases, attempting to more aggressively manipulate the endoscope tip to visualize the ventral commissure can restrict laryngeal motion, which may be mistaken as dysfunction. It may also block laryngeal airflow resulting in asphyxiation.
3. Observe the oscillatory motion of the larynx as the rodent inhales (glottal widening) and exhales (glottal narrowing) during each respiratory cycle.
   NOTE: The rate and magnitude of laryngeal/glottal motion may vary with anesthetic depth; however, a glottal gap (i.e., air space between the left/right arytenoids and vocal folds) typically remains visible throughout the respiratory cycle in healthy rodents.
   1. If marked glottic narrowing is noted, adjust the endoscope position to ensure unobstructed airflow through the upper airway. For example, avoid applying pressure on the velum, which can cause soft tissue obstruction of the nasal airway. Also avoid inserting the endoscope tip into the glottal space, which can block laryngeal airflow leading to asphyxiation. If in rare cases, breathing ceases, deliver several mid-sternal chest compressions (using one finger) or positive pressure ventilation (using a miniature "resuscitation bag") after removing the endoscope.
4. Video record laryngeal respiratory motion for 30-60 s for post hoc evaluation purposes.
5. With the larynx still in close-up view, slightly adjust the endoscope tip within the laryngeal inlet to apply mechanical stimulation to the mucosa overlying the velum and/or epiglottis, and evoke swallowing in optimally anesthetized rodents.
   1. Use micro-level adjustments of the endoscope tip (i.e., ~1 mm in any direction) to prevent mucosal injury and/or airway obstruction.
   2. Watch for evoked swallows, which can be readily identified as abrupt glottic closure events occurring in synchrony with visible jaw depression, tongue EMG bursting activity, and brief (<1/2 s) apnea that is visible in the respiratory trace.
      NOTE: Glottic closure events without co-occurring jaw movement may occur; however, glottic closure is typically incomplete for these cases. We suspect these may be other airway protective reflexes (e.g., laryngeal adductor reflex) emerging as anesthesia is beginning to wear off; however, they are rare/inconsistent occurrences that require further investigation.
   3. Repeat until 5-10 swallows are evoked and video recorded per animal. If swallowing is abolished, remove the endoscope and wait 5-10 min for ketamine metabolism to occur before proceeding.
6. Carefully retract, but do not remove, the endoscope into the oropharynx and center the hypopharynx in the field of view to visualize the epiglottis and velum.
7. Recouple the velum and epiglottis for resumption of nasal breathing by using a micro-swab to apply light pressure against the tongue base to evoke swallowing and re-entrapment of the epiglottis beneath the velar membrane. If recoupling does not occur within a few attempts, proceed with anesthesia recovery without recoupling to avoid risking laryngeal injury.
8. Stop the endoscopy (and electrophysiology) recording.
9. Use a saline-soaked cotton swab to moisten the tongue and central incisors and return the tongue to its anatomical position within the oral cavity.
10. Detach the ear bars and remove the temperature probe, respiratory sensor, and EMG electrodes from the rodent to proceed with anesthesia recovery.

4. Anesthesia recovery

1. Place the animal in a prewarmed recovery cage (i.e., same as the induction cage) on the "warming station" to recover from anesthesia.
2. Reapply eye lubricant to prevent drying.
3. Administer warmed saline SC for fluid hydration: up to 5 mL for rats and up to 0.5 mL for mice.
4. Administer atipamezole SC for xylazine reversal and to increase respiratory drive: 1-2 mg/kg for rats and mice.
   1. Begin with 2 mg/kg atipamezole, immediately followed by manual stimulation along the rodent's back and stomach to accelerate recovery.
      NOTE: Using this approach, spontaneous head movement typically begins within 1-3 min. However, return to ambulatory status typically takes an average of 2 h (ranging from 1 h to 5 h) following laryngoscopy under KX anesthesia, due to individual differences.
   2. Provide additional atipamezole dosed at 1 mg/kg (at least 15 min after the first injection) if spontaneous body movement is diminishing instead of increasing within the first 15-30 min of anesthesia recovery, despite providing frequent manual stimulation.
   3. Proceed with the administration of doxapram (5 mg/kg IP for rats and mice) if spontaneous activity continues to diminish. Re-dose with this "emergency rescue" agent at 10-15-min intervals (up to five doses) until spontaneous movement emerges. If rodents remain moribund, euthanize the animals using approved euthanasia methods (e.g., an overdose of ketamine followed by a secondary method such as decapitation).
5. Closely monitor recovering rodents at 15-20 min intervals to detect adverse changes in respiratory status, mobility, and thermoregulation and provide intervention as needed (e.g., manual stimulation, supplemental oxygen, thermal blanket, atipamezole, or doxapram injections). Provide more frequent monitoring for rodents requiring intervention.
   1. Provide supplemental oxygen (e.g., 1-2 L/min in a warmed induction chamber, without ISO) for 10 min intervals as needed for rodents with protracted KX anesthesia recovery times. Alternatively, apply frequent stimulation along the animal's dorsal and ventral surfaces to normalize SpO₂ levels (>94%).
   2. Place recovering cage mates in the same recovery cage (up to two rodents per cage) to promote increased spontaneous activity and faster recovery.
6. Return rodents to their warmed home cage when able to spontaneously move around the recovery cage.
   1. Return the standard food and water bottle to the home cage. Do not provide special accommodations for food/water access while anesthesia is still wearing off to minimize choking/aspiration risk.
   2. Observe the home cage activity and remove any obstacles that impede ambulation (e.g., hut, PVC pipe).
   3. Position the home cage half on, half off the warming station for the next 12-16 h (i.e., overnight).
      NOTE: Discontinuing supplemental heat earlier may result in mortality due to hypothermia.
7. Perform standard health checks the next morning. Return animals with the resumption of normal/baseline activity, body functions (e.g., thermoregulating, eating, drinking, urinating, defecating), and stable weight (i.e., maintaining or gaining) to standard vivarium conditions with routine daily health monitoring. On the rare occurrence that rodents have diminished activity, body functions, or body weight, continue supplemental heat for another day.
   ​NOTE: For rats, it is common for porphyrin staining to suddenly appear around both eyes ~3-6 h into the anesthesia recovery period. The staining typically resolves within 24 h.

5. Objective quantification of laryngeal motion during breathing versus swallowing

1. Use video editing software with a frame-by-frame analysis feature to view the endoscopy videos.
   1. Identify at least one representative 10-20 s episode of spontaneous breathing per animal.
   2. Identify 3-5 representative swallow events per animal.
   3. Ensure the selected breathing and swallowing episodes/events meet the following analysis criteria: the larynx centered in the camera field of view with all laryngeal structures/borders visible (i.e., not obscured by the velum, epiglottis, or excess salivary secretions), sufficient lighting (i.e., able to see all laryngeal structures/borders), and without camera motion artifact (i.e., the endoscope is not moving).
2. Analyze the identified breathing and swallowing episodes/events using subjective and/or objective approaches.
   1. For subjective analysis: Use a Likert scale to subjectively score laryngeal motion during real-time and frame-by-frame viewing using video editing software. To follow this protocol, use an expanded Likert scale ranging from -2 to +2, where negative values indicate laryngeal motion in the opposite direction than expected. Estimate laryngeal airway protection during swallowing, where 0 = no reduction in the glottal gap size (i.e., no laryngeal airway protection), 1 = some glottal gap reduction (i.e., incomplete airway protection), and 2 = complete adduction of the arytenoids, with only a small ventral glottal gap between the vocal folds (i.e., complete airway protection), with negative values indicating paradoxical laryngeal motion.
      NOTE: A Likert scale ranging from 0 to 2 is commonly used in rodent studies, where 0 = no motion, 1 = some motion, and 2 = normal motion^{17,18,19,20,21,22}. However, this scale does not take into account paradoxical laryngeal motion that often occurs following recurrent laryngeal nerve (RLN) injury¹⁰.
   2. For objective analysis: Identify four key video frames – 1) the rest frame directly preceding initiation of laryngeal adduction (i.e., the frame before the vocal folds adduct), 2) the frame in which the vocal folds have completed adduction, 3) the frame immediately preceding abduction of the vocal folds (this may overlap with #2), and 4) the frame where the vocal folds have abducted completely and are returned to a rest position to resume breathing. Use the time stamp of these four key frames to calculate the duration of vocal fold adduction (frame 1 to frame 2), glottic closure (frame 2 to frame 3), vocal fold abduction (frame 3 to frame 4), and total swallow duration (frame 1 to frame 4). Alternatively, use other existing commercial software²³ to measure the glottal angle (i.e., between the arytenoids dorsally or the vocal folds ventrally) during maximum abduction and maximum adduction using still-frame images^17,18,24. Have at least two trained reviewers perform this process independently in a blinded fashion, identify discrepancies between reviewers, and reach a joint consensus for each discrepancy.
      NOTE: We previously performed this manual frame-by-frame analysis of glottic closure timing events (i.e., during airway protective reflexes) using commercial video editing software in rodents and humans¹⁴. However, this approach is inefficient and provides only limited quantification of laryngeal motion dynamics. We currently use in-house-built laryngeal tracking software to perform more extensive objective quantification of laryngeal motion during breathing and airway protective reflexes^{22,25,26,27,28,29,30}. The software features automated frame-by-frame tracking capability for objective quantification of left- versus right-sided laryngeal motion distance (amplitude) and timing (frequency). We use these parameters to derive a variety of motion-based measurements to detect/quantify laryngeal dysfunction (e.g., glottal angle maximum/minimum/range, laryngeal motion symmetry, and synchrony) during breathing, swallowing, and other airway protective reflexes (e.g., laryngeal adductor reflex). This software is still undergoing refinement and is not yet commercially/publicly available.