A data review of airway management in patients with oral cavity or oropharyngeal cancer: a single-institution experience

In the United States, an estimated 41,000 oral cavity and oropharyngeal malignancies are diagnosed each year [1]. The anatomic sites of oral cavity cancer (OCC) cover the lips, anterior two-thirds of tongue, gingiva, retromolar trigone, buccal mucosa, hard palate and the floor of mouth; whereas, the sites of oropharyngeal cancer (OPC) are located in the posterior one-third of the tongue, palatine or lingual tonsils, soft palate, and posterior pharyngeal wall [2].

With the technologic improvement, radiotherapy, depending the subsite of a disease, has become a primary treatment for many head and neck malignancies in order to maximally maintain functionality of the upper aerodigestive track. Approximately 75% of patients with head and neck squamous cell carcinoma can benefit from head and neck radiation therapy (HNRT) [3, 4]. Although previous receipt of HNRT is a recognized risk factor for difficult airway management, the mechanisms of HNRT- related airway pathologic changes and the overall influences of previous HNRT on the outcomes of tracheal intubation remain unclear.

In this retrospective study, we reviewed the performance and outcomes of tracheal intubation in patients who were undergoing surgery to resect primary OCC or OPC of any stage. The primary goal of the study was to determine if previous HNRT adds additional risk to the airway management in the patients with OCC and OPC disregarding the technique applied. In addition, the factors potentially associated with difficult tracheal intubation (DTI) were also analyzed.

The finding in this study did not reveal the correlation between previous HNRT and DTI in patients with OCC or OPC in spite of significant association between HNRT and restriction of neck range of motion. This can be explained by broadly using flexible endoscopy (35.5%) for tracheal intubation in the studied population, which effectively overcame the influences of restriction of neck range of motion to the performance of tracheal intubation. Nevertheless, our finding does not necessarily mean that the influence of HNRT on tracheal intubation should be underestimated. In this study, 22.3% of patients treated with HNRT had restricted neck movement, whereas only 11.0% of patients in the non-HNRT group did. In addition, although the difference was not statistically significant, more patients in the HNRT group (24.8%) than in the non-HNRT group (18.7%) had trismus. Both restricted neck range of motion and trismus are significant risk factors for difficult direct laryngoscopy. Furthermore, in our clinical experience, trismus and restricted neck range of motion often coexist in patients who develop tissue fibrosis after radiotherapy, posing a significant challenge for both direct laryngoscopy and video laryngoscopy. Because the components of radiation fibrosis syndrome result from the combination of tonic contraction and fibrosis of the muscles of mastication [6], trismus and limited neck range of motion caused by radiation cannot be improved with anesthetic induction or use of a muscle relaxant. This is unlike the trismus induced by cancer pain or inflammation commonly seen in patients who have not undergone radiotherapy, which can often be improved with anesthestics and muscle relaxants. Furthermore, restricted neck range of motion is unlikely to exist in patients without tissue fibrosis unless it is caused by neck-related comorbidities, such as having undergone a neck fusion procedure. No good estimates of the morbidity rate attributable to restriction of neck range of motion after radiotherapy are available, but radiation-induced trismus occurs in up to 45% of patients who receive curative doses of radiation to the head and neck [7]. Thus, having a back-up plan is important when dealing with airways in patients who have undergone HNRT, so that appropriate tools will be available to manage DTI if it arises.

Another significant finding in this study was that MP scores of 3 or 4 were associated with DTI regardless of the technique used. This result is different from a previous report indicating that MP score alone is not significantly correlated with DTI in patients with otherwise normal airway [8]. The discrepancy can likely be attributed to the study populations. The MP scores in our patients reflected the additive effects of oncopathology, radiation therapy and baseline condition. A previous study reported that up to 75% of patients developed edema in the head and neck after radiotherapy, and > 50% of patients had edema in the airway [9]. However, the existing grading systems for assessment of airway edema are unreliable and lack consistency [10, 11]. Thus, anesthesiologists who manage such patients’ airways are unlikely to find clinical reports useful in planning airway management. Patterson et al. [11] developed an airway edema rating scale system by measuring intrarater consistency and interrater agreement on the level of airway edema at several anatomic sites. They found that the aryepiglottic folds and arytenoids were the regions most amenable to visual check (Fiber-laryngoscopy) and that intrarater consistency and interrater agreement were low for the base of the tongue. Because radiotherapy-induced airway edema is mostly diffused throughout the area of treatment, the level of edema found in laryngeal or supraglottic areas may reflect edema in the base of the tongue, pharyngeal well, and floor of the mouth. We speculate that moderate edema in any hypopharyngeal or laryngeal site combined with an MP score of 3 or 4 is highly suggestive of DTI. Therefore, a bedside flexible endoscopy airway assessment before anesthetic induction on a patient suspected of airway edema should be incorporated into the routine practice for the clinicians who manage high-risk airways.

It is worth mentioning that the intubation technique used in any given case is based not only on clinical indications but also on the personal preference and skill level of the provider. Therefore, the techniques used in the study might not always directly reflect clinical indications. Furthermore, the outcomes of intubation of difficult airways in patients with OCC or OPC are highly experience-dependent. The techniques and equipment used for intubation are also influenced by local culture or practice; similar cases may be managed differently in different institutions [12, 13]. In this study, the intubations were all performed or supervised by a group of head and neck anesthesiologists who routinely use the techniques described; thus, the results of this study might not be widely applicable to institutions with different practices.

In addition, although the history of HNRT is important to airway management, the anatomic site of the tumor may also affect the airway management. However, the majority of patients (75%) in this study had tumors with local or regional invasion beyond the primary anatomic sites so that grouping the patients by discrete anatomic margins was not feasible. In addition, these cancers induce tissue reaction or inflammation that often extends the mass effect to a much larger area than the tumor itself. Therefore, the causal relationship between primary cancer site and the airway management was not studied.

In conclusion, history of HNRT does not increase the overall incidences of DTI in patients with OCC or OPC. Mallampati airway assessment may be used as a sensitive tool to predict difficult tracheal intubation in these patients.