Chapter 33. Airway Management in the Operating Room of a Patient with a History of Oral and Cervical Radiation Therapy

A 68-year-old man was found on CT to have a right lung nodule and paratracheal lymphadenopathy. He was then scheduled for diagnostic bronchoscopy and mediastinoscopy.

Twenty years ago, he was diagnosed with carcinoma of the right submandibular gland, and underwent excision of the gland, right radical neck dissection, and a course of radiotherapy. He quit smoking several years ago and has had hypertension for about 5 years. He has had a nonproductive cough for several months. His only medication is metoprolol.

On examination, he is in no distress at rest. His vital signs are: blood pressure (BP) 140/90 mm Hg, heart rate (HR) 70 beats per minute (bpm), and respiratory rate (RR) 18 breaths per minute. Oxygen saturation on room air is 96%. His weight is 94 kg and he is 170 cm tall. Auscultation of the chest reveals decreased breath sounds bilaterally but no rales or rhonchi, and normal heart sounds. No carotid bruits are evident.

Airway examination reveals a Mallampati IV classification. Mouth opening is 2.5 cm and mandibular protrusion is less than 1 cm. Full upper dentition is present but the mandible is edentulous. The thyromental distance is normal. Cervical spine extension is decreased. Palpation of the submandibular tissues (mandibular space) reveals a woody, indurated consistency. On inspection, telangiectasia and pallor of the submandibular skin are noted. The right neck has the typical appearance of a previous neck dissection. The mucosa of the tongue appears dry.

Laboratory data reveal normal electrolytes and a hemoglobin of 140 g·L−1. ECG reveals nonspecific ST and T changes.

The patient has hypertension which is adequately controlled for his surgical procedure. Carcinoma of the lung is suspected on diagnostic imaging. He does not appear to require further medical optimization.

General anesthesia with endotracheal intubation is required for a brief but stimulating surgical procedure.

Radiotherapy inflicts a radiochemical injury to both normal and malignant cells.1 The damage is related to the total radiation dose and the method of radiotherapy delivery. In order to achieve adequate tumor control, damage to normal tissues is inevitable.1,2Acute tissue toxicities from radiotherapy are considered to occur within 90 days of the commencement of treatment, and late effects beyond 90 days of treatment.3 The late effects may not be manifested until years following the radiotherapy.4 In general, tissues with rapidly dividing cell populations such as mucous membranes and skin demonstrate acute effects of radiation (mucositis, desquamation), whereas those with slowly proliferating cells such as connective tissue demonstrate late effects.5 The severity of the late effects of radiation therapy in general cannot be predicted by the severity of the acute effects.5 The mechanism of late tissue toxicity may be parenchymal or stromal cell death, or irradiation injury to the microvasculature.5 Increased vascular permeability leads to deposition of fibrin in the perivascular interstitium and subsequent replacement by collagen.6 An increase in collagen content can be seen as early as 1 week following irradiation.6

Following radiation therapy to the oral cavity, pharynx, or larynx, the mucous membranes can become erythematous within 1 week, and develop areas with white pseudomembranes (mucositis) at about 2 weeks.5 The patches of mucositis may coalesce by the third week.5 This acute mucosal reaction usually heals within 2 to 4 weeks following completion of radiotherapy, although ulceration and necrosis can occur.4 Late effects of radiation on the mucosa are characterized by thinning or atrophy of the epithelium, telangiectasia, dryness, a loss of mucosal mobility, submucosal induration, and occasionally chronic ulceration and necrosis.6 The mucosa is fragile and more susceptible than normal to mechanical injury.5 Edema is seen in the subcutaneous or submucosal soft tissue in the early phase following radiotherapy, can persist for 6 to 12 months,7 and can become chronic.3,8 Fibrosis, one of the most common delayed radiation-associated manifestations, usually appears in subcutaneous tissues within 6 to 12 months of treatment,5 although it can occur as early as 4 to 12 weeks.6 The fibrosis tends to be slowly progressive,5 nonhomogeneous, and variable in extent and severity from site to site.9 The severity of the fibrosis increases when high total doses of radiation and large fraction sizes are used.4 The risk of developing moderate to severe fibrosis has been reported to be about 40%.3 The affected soft tissue loses elasticity and subcutaneous fat6 and is indurated to palpation.1,6 In the presence of moderate to severe fibrosis, contracture of the tissues also occurs.1 In severe cases, the soft tissues develop a woody consistency and may form a hard mass fixed to skin and underlying muscle or bone (Figure 33-1).5 Obstructive lymphedema may also be associated with fibrosis.5 Radiation therapy to the neck can produce a limitation of neck extension (Figure 33-2).6,10 High-dose irradiation of metastatic cervical lymphadenopathy results in more subcutaneous fibrosis in the neck than does a comparable dose in the absence of palpable lymphadenopathy.5

Hypothyroidism occurs in 5% to 10% of patients who undergo irradiation of the lower neck,4 and fibrosis of the apical segments of the lungs can also occur.5

Voluntary muscle exposed to high-dose irradiation can also develop fibrosis, and when the muscles of mastication (the temporalis, masseter, and pterygoid muscles) are involved, trismus can be produced (Figure 33-3).5 Trismus can be seen following radiotherapy for carcinoma of the nasopharynx or oropharynx. The temporomandibular joint itself is however relatively resistant to ankylosis secondary to radiation injury.5 Fibrosis of the pharyngeal musculature can produce swallowing dysfunction11 and a predisposition to aspiration.2 Stenosis of the pharynx or supraglottic larynx can occur and lead to airway compromise (Figures 33-4, 33-5, 33-6, 33-7, 33-8, 33-9, 33-10).5

Laryngeal cartilage covered by normal mucous membrane usually tolerates conventional fractionated high-dose radiation therapy.4,5 However, arytenoid edema, chrondritis, vocal cord palsy,2 and rarely, chondronecrosis can occur (Figures 33-4[B], 33-5[B], 33-7, 33-8[B], 33-9[B], and 33-10).1 Laryngeal edema can occur at any time following the completion of radiation therapy12 and can produce airway compromise.8 Laryngeal chondronecrosis has been reported to occur up to 22 years following radiotherapy.8

Radiation therapy also can produce vascular injury which includes intimal thickening, fragmentation of the internal elastic membrane, atheroma formation, and fibrosis of the media and adventitia.5 A reduction in the microvascular network can ultimately lead to ischemia, and narrowing or obstruction of larger arteries can occur, as can occlusive thrombosis.9 The changes in the vessel walls are similar to those associated with artherosclerosis due to aging.6 Symptomatic carotid atherosclerosis can be a result of cervical irradiation and may require surgical intervention.12

Radiation injury to the salivary glands produces a decrease in saliva production and a change in the composition of saliva.5 Typically, about 60% to 65% of the total salivary volume is produced by the parotid glands, 20% to 30% by the submandibular glands, and 2% to 5% by the sublingual glands.5 The remainder of the salivary volume is produced by anonymous minor salivary glands distributed throughout the oral cavity and pharynx and which are variable from patient to patient.5 The degree of salivary gland dysfunction depends on the volume of the salivary glands included in the radiation field and the total dose administered.5 It is usually not possible to irradiate the pharynx or the upper jugular nodes without irradiating the submandibular glands; however, the parotid and submandibular glands can be partially shielded during treatment.4 A significant reduction in salivary flow occurs within 1 week of fractionated radiotherapy to the head and neck.5 Salivary flow may become barely measurable by the end of a 6- to 8-week course of treatment and the xerostomia may be permanent.6 Xerostomia causes discomfort, alters taste acuity, and contributes to a deterioration in dental hygiene because the tissues become tender.5 The diminished salivary flow has an altered electrolyte content and reduced pH, and promotes dental decay as the normal oral microflora is altered to a highly cariogenic microbial population.5 In the absence of stringent measures to protect the teeth, caries can develop within 3 to 6 months and lead to complete destruction of the dentition within 3 to 5 years.5 Dental extractions from an irradiated mandible can precipitate osteoradionecrosis.4

The patient presented here had palpable fibrosis of the submandibular tissues, decreased cervical extension, trismus, complete loss of mandibular dentition, and a dry mouth.

Radiotherapy to the oral cavity, pharynx, larynx, or neck can produce limited mouth opening, limited cervical spine extension, and noncompliant immobile fibrotic soft tissue in the floor of the mouth and pharynx, as well as alteration of laryngeal anatomy. Airway management can be difficult in the presence of these anatomic changes. The degree of difficulty is dependent on the site and extent of the altered anatomy.

In a review of 53,041 general anesthetics in which mask ventilation had been attempted, Kheterpal et al identified 77 cases of impossible mask ventilation (0.15%).13 Of the subgroup of 310 patients with neck radiation, 3 could not be ventilated. Both univariate and multivariate analyses demonstrated neck radiation to be the most significant clinical predictor of impossible mask ventilation in this patient dataset. Of the 77 patients who were impossible to mask ventilate, 19 (25%) also demonstrated difficult intubation. However, the incidence of difficult intubation in the subgroup of patients with neck radiation was not provided.13

Giraud et al reported face mask ventilation to be easy after induction of general anesthesia in nine patients after oral or cervical radiation.10 LMA placement was often difficult but was successful in all five patients who had received oral radiotherapy, and ventilation was satisfactory.10 Two patients required lateral introduction of the LMA due to limitation of mouth opening.10 LMA placement was easy in the four patients who had received cervical radiation, but positive-pressure ventilation was difficult.10 On fiberoptic examination through the LMA, the vocal cords could not be visualized in any of these four patients due to vestibular-fold collapse. A large epiglottis was also seen in two of these patients. Muscle relaxation did not improve the laryngeal view. Ventilation was impossible in two of the four patients; however orotracheal intubation was successful. Fiberoptic intubation via the LMA was not attempted as the glottis could not be visualized. The authors theorized that the presence of the LMA in a narrowed, nondistensible hypopharynx may have compressed the larynx and thereby produced glottic collapse.10

Ferson et al reported the use of the Intubating LMA-Fastrach (ILMA) in 254 patients with difficult-to-manage airways of whom 40 had airway changes related to previous surgery, radiation therapy, or both.14 In this subset of patients, the authors reported that correct positioning of the ILMA was more difficult, and 10 patients required the use of a smaller ILMA than that indicated by the patient's height and weight. There were no failures to insert the ILMA and ventilation was possible in all cases. Fiberoptically guided intubation through the ILMA was also successful in all 40 patients. The authors felt that a loss of elasticity due to fibrosis in the neck tissues caused positioning of the ILMA to be more difficult and suggested that fiberoptic guidance be used when attempting intubation through the ILMA in this group of patients.14

Langeron et al compared the efficacy of blind intubation through the ILMA with fiberoptic intubation in a group of 100 patients with anticipated difficult intubation undergoing scheduled surgery.15 In this prospective randomized crossover study, following the induction of general anesthesia, intubation was initially attempted by blind intubation through the ILMA or fiberoptic intubation through an Ovassapian airway. In the event of failure of the first technique, the alternative technique was utilized. The first randomly assigned technique failed in seven patients, four in the fiberoptic group and three in the ILMA group. All were successfully intubated by the alternative technique. In the ILMA group, all three failures occurred in patients who had undergone previous cervical radiotherapy scheduled for ENT cancer surgery. In these patients, ventilation through the ILMA was not optimal for performing blind intubation, alignment of the ILMA was difficult, and increased leaks occurred during ventilation although oxygen desaturation did not occur. The authors concluded that the use of the ILMA could not be recommended in patients with previous cervical radiotherapy.15

Following cervical radiotherapy then, mask ventilation may be impossible13 and the use of an LMA for airway management may not be successful due to obstruction at the level of the larynx.10 Positioning of the ILMA14,15 and ventilation through the ILMA may also be more difficult.15 Laryngeal obstruction would also preclude ventilation using other extraglottic devices such as the Combitube™.

Reduced mouth opening due to fibrosis of the muscles of mastication, reduced cervical spine extension, and fibrosis of the structures of the floor of the mouth can make visualization of the glottis by direct laryngoscopy difficult or impossible. Fibrotic subcutaneous and submucosal soft tissues lack compliance and may constitute a poorly mobile woody mass that cannot be elevated easily, if at all, on direct laryngoscopy. Post-irradiation atrophic mucosa is also easily traumatized and bleeding can readily occur. A thickened edematous epiglottis can obscure glottic visualization, and decreased vocal cord mobility may interfere with glottic cannulation (Figures 33-5[B], 33-7, and 33-9[B]). In the literature search performed for this chapter, no data on the incidence of difficult direct laryngoscopy associated with cervical radiation was found.

Tomioka et al reported ventilation by face mask under general anesthesia to be easy in a patient who had undergone radiotherapy for a pharyngeal tumor.16 However, intubation with a #7.0 endotracheal tube was not possible due to tracheal stenosis, which Tomioka et al postulated, may have been produced by the radiation therapy.16

Yancey reported a Cormack/Lehane Grade III view with a #3 MacIntosh blade following induction of general anesthesia in a patient who had undergone left radical neck dissection and postoperative radiation 10 years previously.17 A frozen larynx that was fibrotic and swollen was described. Intubation with a #3 Miller blade and a #6.5 endotracheal tube was difficult but successful.17

Alternative intubation techniques can also be more difficult following radiation-induced changes to the upper airway. The light-guided technique using a lightwand is best avoided in the presence of anatomic distortion of the airway.18 Retrograde intubation may be feasible although laryngotracheal abnormality has been cited as a relative contraindication to this technique as well.19 Limited mouth opening may preclude rigid fiberoptic techniques, and flexible fiberoptic intubation under general anesthesia may be more difficult in the presence of distorted anatomy and decreased mobility of the airway structures. Langeron et al reported failed blind intubation through the ILMA in patients who had cervical radiation.15 However, Ferson et al reported successful fiberoptic intubation through the ILMA in patients who had airway changes secondary to radiotherapy.14 No information on the efficacy of rigid fiberoptic intubation was found in the literature search performed for this chapter.

Subcutaneous fibrosis in the neck can obscure surface anatomical landmarks, obliterate tissue planes, and make a surgical approach to the airway technically challenging (Figure 33-2). Percutaneous cricothyrotomy may fail in this setting, and tracheotomy may require more time to complete and be associated with more bleeding.20

Neck radiation produces a wide spectrum of pathophysiology 13 and airway management of the patient following cervical or oral radiotherapy requires a careful airway assessment. The assessment should focus on the four dimensions of airway management as outlined by Murphy et al, and be used to predict potential difficulty with (1) face mask ventilation, (2) ventilation using an extraglottic device, (3) tracheal intubation, and (4) surgical access to the airway.21

If mask-ventilation and intubation are both predicted to be difficult after radiotherapy, then the airway should be secured with the patient awake. Should general anesthesia be induced in this setting and mask-ventilation be inadequate, rescue ventilation by means of an LMA may also fail.10 The use of an LMA after cervical radiotherapy may in fact be contraindicated.10 Furthermore, rescue by means of a surgical airway may also be difficult.13,20 Topical anesthesia of the upper airway has been reported to produce transient glottic obstruction resulting in a profound reduction in maximum inspiratory and expiratory flows in some normal subjects,22 and in the presence of preexisting airway compromise an increase in resistance to gas flow may be poorly tolerated.23 Complete obstruction has been reported during topicalization and instrumentation of the airway.20,24,25 However, awake fiberoptic intubation, in general, maintains a wide margin of safety and has been said to be the recommended method in the patient with predicted difficult intubation post radiotherapy. Nonetheless, extreme caution must be exercised in the presence of severe airway obstruction if complete obstruction is to be avoided.10,26 Following radiotherapy to the floor of the mouth and/or pharynx, severe trismus may preclude oral intubation techniques, and in this setting awake nasal fiberoptic intubation may be the most reasonable alternative.

If mask-ventilation is predicted to be easy but intubation difficult, then fiberoptic intubation under general anesthesia may be considered, although anatomic distortion and decreased tissue mobility can make fiberoptic visualization more difficult. Fiberoptic intubation via an ILMA has been successful in the presence of radiotherapy-induced changes to the airway.14 However, failure of blind intubation through the ILMA in this setting has been reported.15

In the presence of anatomic distortion but when intubation is predicted to be possible, inhalation induction of general anesthesia may be a reasonable option.23,27,28 When an adequate depth of general anesthesia has been achieved, intubation can be performed during spontaneous ventilation utilizing a curved- or straight-blade laryngoscope, an operating laryngoscope such as the tubular Lindholm scope, or a rigid bronchoscope.23,27 If intubation is not possible, tracheotomy can be performed under general anesthesia. Inhalation induction in the presence of airway compromise can however be difficult.23,28 Complete airway obstruction can occur and an emergency surgical airway may be required.23,28 Meticulous attention to detail is necessary, in particular the maintenance of spontaneous ventilation and the avoidance of airway instrumentation until an adequate depth of general anesthesia is achieved.23 The use of a nasal airway to alleviate obstruction at the level of the soft palate during inhalation induction may be helpful.23

Patients who have an extremely compromised airway, severe stridor, gross anatomic distortion, or a larynx that cannot be visualized on endoscopy should undergo awake tracheotomy performed under local anesthesia.23,27,28

An anesthetic record from 2 years prior to this admission was available for review. Face mask ventilation had been recorded as moderately easy and intubation had been achieved using a lightwand on the third attempt.

Difficult direct laryngoscopy was predicted based on the examination of the airway (Mallampati IV, reduced mouth opening, limited mandibular protrusion, decreased cervical extension, woody induration involving the mandibular space). The predicted ease of face mask ventilation was also uncertain due to the presence of the submandibular induration and limited cervical extension. The recorded experience with face mask ventilation at the previous surgery is not reassuring. Ventilation by means of an LMA or other extraglottic device may be difficult as well in the presence of the existing anatomic distortion. Surgical access to the airway was not predicted to be difficult. An awake fiberoptic intubation was planned.

Routine monitors were attached and IV access established. No sedation was administered. The patient gargled and then expectorated 30 mL of 4% lidocaine. The Devilbiss atomizer was then used to administer 12 mL of 3% lidocaine aerosolized into the right nostril and the mouth. Five percent lidocaine paste was applied to the posterior one-third of the tongue, and an internal approach superior laryngeal nerve block was performed using Jackson forceps and cotton pledgets soaked in 4% lidocaine. The adult bronchoscope was then easily passed through the mouth into the trachea with the patient in the sitting position and using gentle tongue traction. A #8.5 endotracheal tube was then passed easily over the bronchoscope during maximum inspiration to widely abduct the vocal cords. The patient tolerated the intubation well. General anesthesia was then induced and bronchoscopy and mediastinoscopy performed uneventfully.

Severe postextubation laryngeal obstruction due to laryngeal edema has been reported following hepatic resection in a patient who had previously undergone bilateral modified radical neck dissection and radiation therapy.29 However, laryngeal edema was judged to be unlikely following this relatively brief surgical procedure in which the volume of IV fluid administered was small. Furthermore, no evidence of airway obstruction existed preoperatively.

The patient was therefore extubated fully awake in the semi-sitting position in the operating room immediately following surgery. The postoperative course was uneventful.

Radiotherapy to the head and neck can produce limited mouth opening, limited cervical spine extension, noncompliant fibrotic soft tissue in the floor of the mouth and pharynx, and alteration of laryngeal anatomy. Airway management of the patient following cervical or oral radiotherapy therefore requires a careful airway assessment focused on the prediction of difficult mask ventilation, difficult extraglottic device utilization, difficult intubation, and difficult surgical airway. While complete obstruction has been reported during topicalization and instrumentation of the airway, awake fiberoptic intubation, in general, maintains a wide margin of safety and is the preferred method in the patient with predicted difficult mask ventilation and difficult laryngoscopic intubation post radiotherapy. Patients who have an extremely compromised airway, severe stridor, gross anatomic distortion, or a larynx that cannot be visualized on endoscopy should undergo awake tracheotomy performed under local anesthesia.