A pilot study to determine the effects of nasal co-phenylcaine on drug-induced sleep endoscopy

Obstructive sleep apnoea (OSA) is the most common sleep-disordered breathing (SDB) disease encountered with a prevalence of 3–17% in the adult population and is associated with an increasing rate of morbidity and mortality [1, 2]. Drug-induced sleep endoscopy (DISE) is a widespread procedure and is considered today as the most common diagnostic procedure for upper airway (UA) endoscopic evaluation for snoring and obstructive sleep apnoea (OSA).

Since its first description in 1991, DISE has gained increasing popularity, and evidence reports a good correlation between DISE findings and treatment outcomes [3, 4]. UA evaluation in awake OSA patients has limited usefulness because the patterns of UA collapse can change dramatically between awake and asleep, mostly due to differences in the muscle tone [5]. There remain controversies on how to perform sedation in DISE with an increasing need to reduce the amount of sedation infused and improve patient safety. There is an argument that local anaesthesia and nasal decongestant may reduce the inconvenience linked with scope insertion during DISE thus diminishing the total amount of sedation used. Given the known effects of nasal obstruction on SDB [6], nasal decongestant would also improve nasal evaluation during DISE by facilitating nasal endoscopic visibility and eliminating the rhinitis/congestion component of nasal blockage. On this regard, we have demonstrated that complete visualisation of both middle turbinates on anterior rhinoscopy predicts a good nasal airway and can be graded accordingly [7]. However, the European position papers do not recommend the use of nasal decongestants and nasal anaesthetics during DISE owing to the hypothesis that they may affect the grading score outcome [8, 9].

Our aim is to evaluate the effect of co-phenylcaine (lidocaine 5% and phenylephrine 0.5%) administered via nasal spray on 27 patients undergoing DISE for SDB disease and to evaluate if the application of co-phenylcaine could or not significantly affect the grading outcome and thereby influence the patient’s treatment plan.

Clinical and demographic data are summarized in Table 1. Among the 27 cases, 17 did not have a change in DISE outcome, eight showed a change in percentage of contribution to upper airway collapse and two patients showed a change in total DISE grading. Findings observed at the examination are reported in Table 2.

Considering UA collapse, the most frequent UA collapse grade recorded in our population was Grade 3, both before and after nasal spray application. Upon first performing DISE, 1 patient (3.7%) had grade 2 UA collapse, 20 patients (74.1%) had grade 3 and 6 patients (22.2%) had grade 5 UA collapse. After nasal spray administration 22 patients (81.5%) had a grade 3 and 5 patients (18.5%) a grade 5 of UA collapse. (Fig. 1) Analysing the findings obtained after spray administration, 17 patients (63%) showed no changes in terms of UA collapse grade or of percentage in level contribution (in case of a multi-level collapse). We observed a change in the percentage of level contribution (without a change in the grade) in only 8 patients (29.6%), all of them with a grade 3 UA collapse. Of these patients, one of them had an increase in velopharynx percentage contribution, two in the oropharynx, one in both the velopharynx and the oropharynx, and four showed an increase in the tongue base percentage contribution. (Fig. 2) However, change in DISE total grading after nasal spray administration was not statistically significant (p = 1.000 with effective Spearman pairing r = 0.841). Co-phenylcaine nasal spray did not change the total Kotecha/Lechner grading system with a statistical power (1 − β) of 0.92. For the same grading, changes in percentage of contribution to collapse were not statistically significant (p = 0.2702 for velopharynx, p = 0.5261 for oropharynx and p = 0.6089 for tongue base). Interestingly, in two patients (7.4%) we observed a change in the grade of UA collapse. (Fig. 2) In particular, one patient had first a grade of 5 and improved to a grade 3, and the other was initially grade 2 and then changed to grade 3 after spray administration.

We could not find any significant factor that could influence the effect of co-phenylcaine on DISE outcome, loudness of snoring or management plan. Nasal spray application did not affect the results of DISE independent of sex, AHI, BMI, tonsils grade, presence of rhinitis, turbinate hypertrophy, nasal septal deviation, or NPIF limitation.

Our results demonstrate that the use of co-phenylcaine does not significantly influence the DISE outcome and importantly does not change the patient’s management plan. Nasal spray application did not affect the results of DISE independent of the gender, AHI, BMI, tonsil grade, presence of rhinitis, turbinate hypertrophy, nasal septal deviation, or NPIF limitation.

According to the position papers on DISE, the use of local anaesthetic during the procedure is not recommended due to the possibility of having an influence on the tone of the pharyngeal muscles of the UA [8, 9]. The exact mechanism through which nasal local anaesthesia may interfere with palatal and pharyngeal muscles is not completely clear. Recent information suggests that the mechanism inducing arousal from sleep during airway occlusion in patients with OSA can be controlled by a complex evaluation made by the central nervous system of both the chemoreceptor stimulation (hypoxia and hypercapnia), the level of effort based on the level of ventilatory drive and/or the feedback from mechanoreceptors in the respiratory muscles or in the UA. Nevertheless, the contribution of UA mechanoreceptors to the arousal stimulus has not been clearly demonstrated. Topical nasal anaesthetic could increase apnoea duration by reducing the input from mechanoreceptors in the UA, even if results reported in the literature on this regard are conflicting. Basner and co-workers reported an increase in the time to arousal from non-rapid eye movement after application of 4% lidocaine to the oral and nasal mucosa [14]. Similarly, Berry et al. showed that topical UA anaesthesia increased the duration of obstruction apnoeas and the levels of inspiratory effort (oesophageal pressure deflection) prior to arousal from non-REM sleep [15]. Conversely, Redolfi et al. reported that topical lidocaine sprayed in the nasal and buccal cavities had no effects on respiratory-related evoked potentials [16]. Nasal negative pressure receptors may mediate activation of UA dilator muscles in the presence of nasal inspiratory negative pressure and/or airflow, and selective anaesthesia of the nasal mucosa is able to cause a decreased activation of the genioglossus and alae nasi muscles [17‐19]. In addition, the effect of topical anaesthesia on UA receptors has also been investigated after topical oropharyngeal application of anaesthetic, but the results are contradictory [20‐22]. Interestingly, Deegan and colleagues hypothesized that UA protective reflexes might not be important in patients with OSA and that these reflexes might already be significantly impaired in these patients, suggesting that oropharyngeal anaesthesia might have little or no additive effect in this category of patients [22].

Even in the case of nasal decongestant, the position papers do not advocate its use as it may interfere with the nasal resistance, thus influencing the airflow and the dynamics of the UA [8, 9]. Specifically, in patients with SBD disease, an increased nasal resistance results in UA closure in snorers [23] and OSA patients [24]. In addition, the volume of the pharyngeal mucosa plays an important role in controlling UA diameter, by maintaining pharyngeal patency, and topical application of phenylephrine in the UA resulted in a decrease in mucosal water content and a consequent increase in the pharyngeal cross-sectional area [25]. On this regard, Hutt et al. reported a significant reduction in SDB in patients with OSA after oropharyngeal and nasopharyngeal topical application of oxymetazoline [26]. It has also been proposed that the instillation of phenylephrine in the UA, by causing a reduction in the UA calibre, may reflexly activate some afferent fibres responding to changes in the size of the pharyngeal airway, finally leading to a further decrease in UA resistance [27‐29]. Conversely, Wasicko et al. found no changes in UA muscle activity after pharyngeal and nasal plus pharyngeal instillation of phenylephrine [30] and, more recently, Clarenbach et al. concluded that the efficacy of nasal decongestion is not sufficient to provide a clinically substantial improvement of OSA, even if a reduction of the apnoea/hypopnoea index was observed [31].

In our department, co-phenylcaine is commonly administered prior to a nasal endoscopy to reduce the unpleasantness linked with the scope insertion, due to the high sensitivity of the nasal mucosa. Considering the advantages linked to the use of nasal anaesthesia and decongestant, as already mentioned, we decided to investigate if their use during DISE could modify the findings observed before their administration. In our study, co-phenylcaine nasal spray caused a modification in the DISE findings in 10 patients (37%). In particular, in 8 of them (29.6%) this modification consisted in a change in the percentage of level contribution, even if it was not so relevant to cause a plan modification. Furthermore, only in 2 patients (7.4%) this alteration implicated a change in the grade of UA collapse and a consequent modification of the surgical plan.

Contrary to our expectations, we observed in 13 patients (48%) an increase in snoring after nasal spray administration. In fact, it can be speculated that, as nasal decongestion increases nasal airway volume, patients should experience a higher nasal airflow, thus leading to an opening of UA and a consequent decrease of snoring. However, our findings are in line with previous studies which suggest that a reduction in nasal resistance during sleep may not correlate with snoring [32‐34]. Nonetheless, we were unable to find a statistically significant relationship between spray administration and snoring.

The nose plays an important role in the development of SDB as nasal breathing is physiologically the preferential breathing route in wakefulness and in sleep [35]. Nasal obstruction in normal volunteers markedly increases the number of obstructive apnoeas and hypopneas during sleep [36‐38]. In addition, patients with symptoms of rhinitis or with nasal congestion are at higher risk of developing snoring or OSA [39, 40] and a correlation has been found between total nasal resistance, AHI and oxygen desaturation [41]. Interestingly, in our study odds ratio calculation showed no influence of rhinitis, turbinate hypertrophy, nasal septal deviation or nasal obstruction (low NPIF) on the changes observed in DISE after spray application as well as the increase in the loudness of snoring. These results may suggest that improvement of nasal resistances, by means of nasal decongestants, may not influence UA collapse and snoring. A hypothesis that may explain our results is that oral breathing could be the preferred breathing route not only in subjects with a chronic nasal obstruction [36‐40] but also in chronic apnoeics and snorers without a nasal obstruction [42, 43]. Thus, an improvement in nasal resistances may be irrelevant in improving UA collapse or decreasing snoring, at least in the short term.

A limitation of our study was the lack of a tool to assess the depth of sedation and anaesthesia, [e.g. bispectral (BIS) index or cerebral state index (CSI)]. It can be argued that, due to the absence of this tool, we cannot be sure if the changes observed 10 min after co-phenylcaine application were a consequence of the spray administered or of the deepening of the sedation. However, our experience is based on thousands of DISE procedures performed each year and sedation is administered by high-skilled anaesthetists in this procedure. In addition, as we used a spray made by a combination of lidocaine and phenylephrine, we were not able to distinguish if the results observed were caused by one of the two drugs or by a combination of both.

A further limitation of our study was that a post decongestant NPIF was not performed as this would have allowed us to potentially determine the NPIF diagnostic percentage change and help differentiate between congestable and structural causes for nasal obstruction [44]. Overall, there is a need to improve the rhinological assessment component during DISE. However, our pilot study only recruited a small number of patients and although the evidence is supportive there is a need for a larger study to fully evaluate the benefits of using co-phenylcaine during DISE and the need to expand on the rhinological assessment of SDB.

Our pilot study supports the use of co-phenylcaine nasal spray during DISE. The advantages of using a nasal decongestant include improving nasal endoscopic visibility through the elimination of nasal congestion as well as aiding in the diagnosis of congestible versus structural causes of nasal congestion. The anaesthetic component could also lower the total dose of intravenous anaesthetic administered during the procedure by reducing patient discomfort.