Feasibility of face mask spirometry during decannulation in head and neck surgery: Prospective cohort study

Abstract Objectives To analyse the relationship between spirometric parameters measured with a face mask versus a mouthpiece, as well as the feasibility of face mask spirometric evaluation in a head and neck surgery (HNS) decannulation context. Furthermore, we examine peak inspiratory flow (PIF) cut‐off values before and after decannulation. Design Prospective cohort study. Setting Otolaryngology HNS Department of a university teaching hospital. Participants Twenty‐four patients were selected. A maximal flow‐volume loop was conducted before (with mouthpiece) and after (with mouthpiece and face mask) decannulation. Main outcome measures Recorded outcomes were forced vital capacity (FVC), forced expiratory volume in the first second, peak expiratory flow, PIF, forced expiratory flow at 50% of FVC and forced inspiratory flow at 50% of FVC. Spearman correlation coefficients between spirometric parameters measured with a face mask versus a mouthpiece were calculated. Wilcoxon test was used to check differences between mouthpiece and face mask values. Results Correlation between mouthpiece and face mask spirometric values was moderate to high (r = 0.46–0.95). All parameters measured by spirometry were significantly lower with a face mask than those obtained with a mouthpiece (p < 0.05). Before decannulation, the lowest PIF value (tested with mouthpiece) that allowed successful decannulation was 1 L/s. After decannulation, the lowest PIF value tested with mouthpiece and face mask for successful completion of the decannulation process were 0.77 and 0.56 L/s, respectively. Conclusion Face mask is a feasible option to perform a spirometry when face diseases hinder spirometric evaluation through a mouthpiece in an HNC surgery context.


| OBJECTIVES
Temporary tracheostomy is commonly used after head and neck surgery in order to prevent severe upper airway obstruction (UAO) consequences, swallowing dysfunction or difficulties in managing secretions. However, tracheostomy remains a procedure associated with acute and late complications. 1 In addition, tracheostomised patients need close monitoring and specialised care. Thus, as soon as physiological breathing is possible, decannulation is a desirable but challenging process.
Capping trial is the most used method prior to cannula removal.
In this respect, published protocols recommend that decannulation is warranted after successful capping for 24-48 h. 2 In recent years, a more functional approach to assess UAO with spirometry has been used at the bedside and in ambulatory patients before decannulation in head and neck cancer (HNC) patients after surgery. 3,4 Although some guidelines can be found in the literature, 3,4 spirometry cut-off values must be optimised to decrease decannulation failure rates.
On the other hand, there are certain contexts such as limited mouth opening or neurological oral sphincter incompetence that prevent spirometric evaluation through a mouthpiece. Limited mouth opening after surgery can be the result of local resection, oedema, pain, temporo-mandibular disorders or radiotherapy sequelae. Furthermore, neurological oral sphincter incompetence may be the consequence of transient or permanent damage of perioral nerves during surgery. Accordingly, face mask (FM) spirometry may be a good strategy to evaluate patients with these orofacial disorders. In this regard, the updated ATS/ERS spirometry guideline considers the use of FM in patients unable to use a mouthpiece. 5 Until now, studies related to FM use focused their attention mainly on peak cough flow (in order to establish cough efficacy), forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV 1 ). [6][7][8][9] Peak inspiratory flow rate (PIF) has not been addressed.
However, this parameter is currently used to evaluate upper airway obstruction and/or basis to determine the time of decannulation of patients after HNC surgery. 3,4,[10][11][12][13] The aim of this study is to analyse the relationship between FM versus a mouthpiece spirometric values, as well as the feasibility of FM spirometric evaluation. Furthermore, we examine PIF cut-off values before and after decannulation with mouthpiece and FM in an HNC surgery context.

| DESIGN
A prospective cohort study was designed and conducted.

| SETTINGS
The study was performed between November 2017 and January

| PARTICIPANTS
A total of 25 consecutive patients were eligible. Inclusion criteria were tracheostomised adults following elective surgery for HNC. Some of these patients stayed in the intensive care unit for a few days in order to monitor the free flap performed during surgery before discharge to HNC Surgery unit. Patients who failed to achieve at least three valid measurements before and after decannulation with mouthpiece and FM were excluded. clip. After decannulation, measurements were performed with the mouthpiece (with a nose clip) and with a FM (Intersurgical Limited) connected to a sectioned mouthpiece (48 tests in all). A minimum of three FVLs were performed in each case before and after decannulation. For statistical analysis, FVLs with the best summative PEF-PIF were selected. 4

| Decannulation and spirometric test protocols
Patients were tracheostomised with a non-fenestrated cuffed tube (Shiley; Covidien plc). The cannula was usually changed the day after surgery for a fenestrated cuff-less tracheostomy tube of the same diameter. Decannulation was commonly scheduled for the 3rd-6th day after surgery. The decision to decannulate was made by the physician following physical examination since the 3rd day after surgery, provided the PIF values were at least 1 L/s with the cannula still inserted. Decannulation was postponed in the presence of a medical complication (heart or respiratory failure, pneumonia or neurological complication) or early locoregional complications (haematoma or infections). In some cases, when decannulation could not be performed within the expected timeframes or when a maladjusted tracheostomy cannula was suspected, verification by fiberoscopy and/or spirometric assessment without cannula were also performed.
Decannulation failure was defined as the need to recannulate the patient within 48 h of decannulation, whereas decannulation success was defined as the ability to continue breath without a tracheostomy with no significant dyspnoea.
Before decannulation, the FVC manoeuvre was obtained through the patient's mouth with the cannula (including the inner cannula) still inserted (and by obstructing the fenestrated cuff-less tracheostomy tube) with a mouthpiece and a nasal clip. After decannulation, the FVC manoeuvre was obtained through a mouthpiece and FM (in both cases by obstructing the tracheostomy orifice). For FM spirometric tests, the patients were instructed to put the FM on their face and to perform the spirometric test to avoid air leaks. Decannulation and spirometry procedures were closely monitored by the same physical therapist, for each patient and with different time measurements.
Moreover, to avoid the effects of fatigue or learning effects, after decannulation, the FM and mouthpiece spirometric tests were conducted in a random order.

| Study size
The study population was a consecutive series of 24 patients who met the inclusion criteria and signed written informed consent. The study size had not been planned ahead. In any case, the recruitment process was declared completed prior to statistical analysis. 6 | RESULTS

| Sample characteristics
Twenty-five tracheostomised patients after HNC surgery were eligible during the study period ( Figure 1). One patient was unable to give informed consent, consequently, 24 patients were finally analysed. 6.3 | Relationship between spirometric parameters before and after decannulation obtained with a mouthpiece Correlation between PIF, FIF 50%, FEV 1 and FVC values before and after decannulation was moderate to high (r = 0.575-0.81). Nevertheless, the correlation between FVC/FEV 1 , PEF and FEF 50% values before and after decannulation could not be demonstrated (p > 0.05).

T A B L E 2 Spirometric parameter values after decannulation obtained with a mouthpiece and a face mask (n = 24)
All parameters measured were significantly higher after decannulation (p < 0.05), except FEV 1 /FVC ratio and FIF 50% (Table 3).
6.4 | Peak inspiratory flow cut-off value with cannula for determining decannulation (measured through a mouthpiece) Median PIF value measured through a mouthpiece before decannulation was 1.57 L/s. The lowest PIF value before decannulation was 1.01 L/s.
For all other parameters, see Table 2. Figure 2 shows the behaviour of PEF and PIF values before and after decannulation.
F I G U R E 2 Peak expiratory flow (PEF, L/ s) and peak inspiratory flow (PIF, L/s) values before and after decannulation using a mouthpiece (n = 24) 6.5 | PIF cut-off value without cannula for determining decannulation (measured through a mouthpiece and a face mask)  Table 3. Figure 3 show the behaviour of PEF and PIF values with both interfaces after decannulation.

| DISCUSSION
This study analysed the use of an FM for the evaluation of the upper airways in the decannulation process of patients in the context of HNC surgery. FM spirometric evaluation was feasible: all patients were easily evaluated with this interface and we cannot report any adverse event or difficulty worth highlighting. In accordance with previous studies, our results show that all spirometric values obtained with a mouthpiece were significantly higher than with an FM (p < 0.05). 7 We can corroborate that the PIF and PEF values also follow this trend ( Figure 3)

| Limitations and strengths
Although in this sample all patients could be decannulated at the first attempt, the reduced size of our study population prevents us from drawing final conclusions about the decannulation cut-off.
A debatable aspect of our protocol is the absence of an FM test before decannulation. We consider that performing a spirometric evaluation with both interfaces before and after decannulation could overload patients.
Despite the small sample of our study, we believe it is representative of HNC patients in terms of age, gender, smoking habits and type of tumour. 18

| CONCLUSION
In HNC patients with limited mouth opening or oral neurological sphincter incompetence after surgery, using an FM to perform a spirometry is both feasible and reliable. A PIF value of 1 L/s or higher, measured through a mouthpiece, was associated with a successful decannulation.

AUTHOR CONTRIBUTIONS
José Antonio Sánchez-Guerrero conceived, designed and performed the study, analysed, wrote, revised and approved the final manuscript.
Maria Àngels Cebrià i Iranzo conceived, analysed, wrote, revised and approved the final manuscript. Francisco José Ferrer-Sargues contributed to the analysis and revised and approved the final manuscript.
Sophie Périé conceived, designed, supervised and revised and approved the final manuscript.