Introduction
Tracheostomy is commonly performed during major head and neck surgery for airway protection. Unlike a nasotracheal or orotracheal tube, a tracheostomy moves the airway circuit away from the operative field, and affords free movement of the neck and jaw without fear of tube displacement. In the post-operative period, a tracheostomy offers airway protection from oedema and haemorrhage, and can be utilised for anaesthesia for any re-exploration. Further, it reduces the need for sedation and intensive care unit support, enabling care to be shifted directly to the ward. Once airway oedema settles, the tube is removed.Reference Castling, Telfer and Avery1, Reference Marsh, Elliott, Anand and Brennan2
A tracheostomy is, however, a double-edged sword. While protecting the airway, it can also cause potentially life-threatening complications such as tube block.Reference Castling, Telfer and Avery1 In this study, we assessed the outcomes of tracheostomy after major oral and oropharyngeal resections. Within this cohort, we compared single cannula versus double cannula tracheostomy tubes with regard to complications such as tube block, and necessity for tube change or re-insertion for continued airway protection.
Materials and methods
Institutional ethics committee approval was obtained for this study. All patients undergoing tracheostomy during major oral or oropharyngeal tumour resection were included. A single cannula or double cannula tracheostomy tube was used, depending on the primary physician's choice. Data were collected prospectively, but analysis was conducted retrospectively. The study period was from 2011 to 2015. Outcomes of patients with a single cannula were compared to those with a double cannula. Only subjects undergoing primary tracheostomy were included, while those undergoing revision tracheostomy were excluded.
Selected patients undergoing major oral or oropharyngeal resection underwent tracheostomy for airway protection. Management protocol for all patients, regardless of the type of tube used, was the same. Antibiotics (amoxicillin/clavulanic acid and metronidazole) were given at the time of induction. Unless a patient had stridor or severe airway narrowing, nasotracheal or orotracheal intubation was carried out.
After neck dissection and before commencement of oral surgery, tracheostomy was performed via a midline 1.5 cm vertical incision in the lower neck. Strap muscles were separated and the thyroid isthmus was divided. Tracheostomy was performed between the third and fourth rings, and an inferiorly based Björk flap was sutured to the skin. A flexometallic endotracheal tube was inserted into the tracheostomy, which was changed to a tracheostomy tube at the end of the procedure. For the single cannula, a cuffed Portex (blue line) tube (Smiths Medical, London, UK) was used, with an inner diameter of 8.5 mm (outer diameter of 11.6 mm) for males and an inner diameter of 8.0 mm (outer diameter of 11.0 mm) for females. For the double cannula, a cuffed Shiley tube (Covidien, Boulder, Colorado, USA) was used, with an inner diameter (inner cannula) of 6.4 mm (outer diameter of 10.8 mm) for both males and females. The tube was stitched to the neck skin.
After observation for 4 hours in the recovery unit, patients were shifted directly to the ward for post-operative care. Unless significant co-morbidity was present, intensive care unit monitoring was not routinely opted for. The subject's head was kept elevated by 45 degrees. Antibiotics were continued for 72 hours and proton pump inhibitor was prescribed for a minimum of 2 weeks. Steroids were not given routinely.
All head and neck surgical patients with a tracheostomy were managed by experienced nurses. All patients received continuous humidified oxygen (4 litres per minute) by mask to the tracheostomy, and suction 4 hourly and as needed. Tube patency was checked at 12 hourly intervals by the otolaryngology service. Intensive care unit respiratory therapists with additional experience and training in handling tracheostomy tube block were available at all times for respiratory emergencies. Any noisy breathing (regardless of tube type) was initially handled by the ward nurse via suction removal of a mucus plug or blood clot with saline irrigation if necessary. If this did not provide clearance of a double cannula, the inner cannula was removed, cleaned thoroughly and re-inserted. If despite these interventions, there continued to be little or no airflow through the tube, or there was respiratory distress as reflected by falling oxygen saturation, tachypnoea or use of accessory muscles of respiration, simultaneous calls were made to the attending physician and respiratory therapist. If tube block was confirmed, the tube was immediately removed.
Chest physiotherapy was commenced twice daily from the 1st post-operative day, and early ambulation was encouraged. From the 2nd post-operative day, the tracheostomy cuff was left deflated and patients were assessed daily for decannulation. If the posterior pharyngeal wall was clearly visible with mouth opening and tongue depression, and nasal fibre-optic assessment revealed adequate oropharyngeal and laryngeal lumen with mobile vocal folds, decannulation was deemed safe. Regardless of the type of tube used, it was removed at this stage and daily occlusal dressing of the tracheocutaneous fistula was started. Patients were monitored in the hospital for a minimum of 24 hours following decannulation. If airway oedema persisted beyond the 4th day, a single cannula was changed to a double cannula. We felt that by the 4th day, the tracheocutaneous tract would be sufficiently mature to allow for safe tube change.
Single cannula and double cannula patient groups were compared with regard to various outcomes. Primary outcomes were: significant obstruction warranting immediate removal of the tracheostomy tube, and the need for a change of tube or tube re-insertion for continued airway protection. Secondary outcomes were: time to decannulation, pulmonary complications, wound infection or dehiscence, duration of hospital stay, time to resume a soft diet, tracheocutaneous fistula healing time, and treatment cost.
Chi-square or Fisher's exact tests were conducted as appropriate for categorical variables, and the Mann–Whitney U test was employed for continuous variables. A two-tailed p value of 0.05 or less was considered statistically significant.
Results
A total of 96 consecutive patients underwent tracheostomy at the time of oral or oropharyngeal tumour resection. Forty-six patients had a single cannula and 50 patients had a double cannula. Patient demographics, tumour characteristics and surgical details are presented in Table I.
Table I Patient, tumour and surgical characteristics
*n = 46; †n = 50. ‡Sarcoma, salivary gland carcinoma, odontogenic keratocyst or retropharyngeal lipoma. **For neck haematoma or thrombosis of free flap. SCC = squamous cell carcinoma; ASA = American Society of Anesthesiologists; COPD = chronic obstructive pulmonary disease
Both the single cannula and double cannula groups were comparable at baseline evaluation. Prior radiation had been given in one and two patients in the single cannula and double cannula groups respectively. Most patients had an advanced stage oral cavity tumour requiring mandibulotomy or segmental mandibulectomy with flap reconstruction. Only eight patients (8 per cent) (four in each group) required elective post-operative stay in the intensive care unit because of chronic obstructive pulmonary disease, cardiovascular disease or long duration of anaesthesia. Most patients were transferred to the ward within 24 to 48 hours. No patient required elective post-operative ventilatory support.
Tube block warranting immediate removal occurred in seven cases (15 per cent) in the single cannula group. Event times were at 38, 44, 48, 48, 50, 72 and 90 hours of the post-operative period. Two of these seven patients had severe respiratory compromise as a result of the tube block, with no airflow through the tube, tachypnoea and oxygen saturation levels below 50 per cent. One patient progressed to apnoea and respiratory arrest before the single cannula was removed and a size 5.0 flexible endotracheal tube was inserted into the tracheostomy site and ventilated. Both patients made immediate and full recovery, with no cardiac or neurological sequelae. Five of these seven patients required immediate tube re-insertion (with a double cannula) because of persistent upper airway oedema.
Seven other patients without blockage of a single cannula underwent elective tube change (to a double cannula) because of persistent airway oedema. Two other patients in whom a single cannula was removed required re-insertion (with a double cannula) within 48 hours: one had fluid overload and pulmonary oedema requiring ventilatory support, and the other had suboptimal work of breathing due to untreated hypothyroidism. Nine patients (20 per cent) therefore required a second tracheostomy tube for continued airway protection or support.
All 16 single cannula patients with primary outcomes were eventually decannulated at a mean post-operative time of 7 (±4 standard deviation) days. The remaining 30 patients (65 per cent) underwent uneventful decannulation of the single cannula on or before the 4th day (Figure 1).
Fig. 1 Flowchart illustrating primary outcome measures. SC = single cannula; DC = double cannula
In the double cannula group, in contrast, none of the 50 patients developed tube block warranting removal (p = 0.004 when compared with the single cannula group). Forty-seven patients underwent uneventful decannulation. One patient underwent decannulation on day 5, but developed stertor and respiratory distress on day 8 due to extensive crusts obstructing the oropharyngeal airway. The (double cannula) tracheostomy tube was re-inserted and he was eventually decannulated on day 10. Finally, decannulation could not be achieved in two patients because of a bulky oropharyngeal flap or prolapse of the tongue base (Figure 1). Excluding these last 2 patients, the need for a second tracheostomy tube in the double cannula group (1 out of 48) was significantly less when compared to the single cannula group (16 out of 46; p = 0.007).
There was no instance of tracheal block due to crusts distal to the tip of the tube in either group. There was no difference in the proportion of patients achieving decannulation within 4 days (65 per cent versus 60 per cent).
There was no difference between the single cannula and double cannula groups with regard to secondary outcomes, including time to decannulate (Table II). There was no case of accidental decannulation or surgical mortality. All patients were followed up for a minimum of four months, during which time no tracheal complications such as stridor developed.
Table II Secondary outcomes
*n = 46; †n = 50. SD = standard deviation
Discussion
The current study had two main objectives. Firstly, to assess the safety of single cannula versus double cannula tubes, and secondly, to evaluate whether tracheostomy was required only for the immediate post-operative period (i.e. within 4 days), or for a longer period (more than 4 days) because of persistent or unanticipated airway pathology. As a double cannula is generally preferred for longer-term tracheostomy,Reference Mitchell, Hussey, Setzen, Jacobs, Nussenbaum and Dawson3 we assessed the proportion of patients in the single cannula and double cannula groups who converted to a double cannula or required another double cannula respectively.
Our data showed that respiratory compromise precipitated by tube block could be avoided by using a double cannula, thus demonstrating its superior safety. There was no instance of blockage warranting tube removal in the double cannula group, whereas seven patients with a single cannula incurred this problem (p = 0.004). The single biggest advantage of a double cannula is the presence of an inner cannula that can be easily removed, cleaned and re-inserted when blocked, while the outer cannula remains in situ, and keeps the airway secure and open. This allows lesser skilled caregivers, including the patient himself or herself, to address tube block in a safe manner. This advantage is lost with a single cannula, as any block that cannot be cleared by suction requires timely recognition of the imperilled airway and tube removal. This risks collapse of a fresh tracheocutaneous tract and further respiratory compromise.Reference Castling, Telfer and Avery1, Reference Halfpenny and McGurk4
To our knowledge, no previous study has compared the use of single cannula with double cannula tubes in head and neck surgery, or any other patient cohort. The safety profile of one tube versus the other is unknown. Important data, such as risk of tube block due to a mucus plug, tracheal obstruction by crusts distal to the tube tip and whether a particular tube provides adequate protection until airway oedema fully resolves, are lacking. Even among studies that focus on complications of tracheostomy in major head and neck surgery, the data on the type of tube used are often absent.Reference Castling, Telfer and Avery1
There are currently no guidelines to help the surgeon choose one tube over another. Therefore, there is wide variation in international practice. While the double cannula has been used in studies from the USAReference Leder, Joe, Ross, Coelho and Mendes5 and Italy,Reference Salgarelli, Collini, Bellini and Cappare6 other centres in Europe and Asia report use of the single cannula in head and neck surgery.Reference Malata, Foo, Simpson and Batchelor7, Reference Sivakumar, Prepageran and Raman8 In several other studies that do not explicitly state the type of tube used, episodes of respiratory arrest due to tube block, similar to the current study, have been reported.Reference Castling, Telfer and Avery1, Reference Halfpenny and McGurk4 It is highly likely that single cannulas have been used in these patients. These findings indicate that the single cannula is preferred for airway protection by many head and neck surgeons worldwide.
There are several reasons for choosing a Portex single cannula over a Shiley double cannula in head and neck surgery. First, the former is less expensive. Second, the single cannula offers the maximum possible inner diameter, which is preferable for elective post-operative ventilation.Reference Mullins, Templer, Kong, Davis and Hinson9 Third, patients may be routinely managed in a post-operative intensive care unit, where one-on-one nursing care reduces the risk of tube block.Reference Marsh, Elliott, Anand and Brennan2 Lastly, post-operative airway oedema might be expected to be short lived, particularly for anterior oral cavity resections. A single cannula might be considered adequate in the anticipation that it could be removed in 2 to 4 days, before problems of crusting and tube block manifest.Reference Castling, Telfer and Avery1
However, recent data show the drawbacks of elective ventilation, even in patients undergoing complex free flap reconstruction.Reference Clemens, Hanson, Rao, Truong, Liu and Yu10 The present study corroborates the benefits and low rate of complications associated with early (‘on-table’) extubation and spontaneous breathing. In this cohort, the slightly reduced inner diameter of the Shiley double cannula tube did not increase the work of breathing. There is also now an increasing trend to shift patients directly to the ward to save costs and reduce nosocomial infections.Reference Arshad, Ozer, Thatcher, Old, Ozer and Agarwal11 However, the lack of one-on-one nursing care on the ward, combined with significant oral or oropharyngeal oedema and oozing of blood from mucosal suture lines increase the risk of respiratory obstruction and tube block caused by thick mucus and blood clot.Reference Castling, Telfer and Avery1, Reference Malata, Foo, Simpson and Batchelor7
• Tracheostomy is commonly performed during major head and neck surgery for airway protection
• However, the safety and adequacy of a given type of tracheostomy tube is unknown
• The present study shows that double cannula tubes offer the best protection and are adequate for the entire duration of tracheostomy requirement
In the present study, 20 per cent of patients in the single cannula group required an elective change to a double cannula because of persistent airway oedema lasting beyond the initial post-operative period. This further illustrates the inadequacy of single cannulas in maintaining a safe airway for the entire duration of tracheostomy requirement. Several factors, such as concurrent neck dissection, previous irradiation and prolonged translaryngeal intubation, can result in persistent airway oedema.Reference Burkle, Walsh, Pryor and Kasperbauer12, Reference Sanguineti, Adapala, Endres, Brack, Fiorino and Sormani13 It is currently not possible to predict when airway oedema would fully resolve in a given patient. In fact, most of the tube blocks in this report occurred within 48 hours, indicating that the single cannula is unsafe even for very short-term needs.
Although the current study was retrospective, management protocol was defined a priori and data were collected prospectively. The same protocol, including that for tracheostomy care, was adhered to for all patients. The single cannula and double cannula groups were comparable at the baseline evaluation. Further, the outcomes studied were robust, well defined, objective, and easily quantifiable and reproducible. Therefore, the potential for bias is minimal. A randomised study might confirm our findings with more statistical validity. However, such a study would raise serious ethical questions given the vastly superior safety profile of the double cannula and the potential for fatal respiratory obstruction with the single cannula as demonstrated here. We believe the data presented here are sufficiently strong to justify exclusive use of the double cannula. This tube is safe, easy to handle and protects the airway until oedema fully resolves (and therefore does not need to be changed).
Conclusion
Double cannulas are safer than single cannulas for airway protection in the tracheostomised head and neck surgical patient. Whether double cannulas other than the Shiley tube, such as the Portex blue line fenestrated, Bivona (Smiths Medical, London, UK) or Duratwix (Boston Medical Products, Westborough, Massachusetts, USA) tubes, are equally safe may be explored in future studies.
