Driving standards in tracheostomy care: a preliminary communication of the St Mary’s ENT-led multi disciplinary team approach


Roland Hettige, Department of Otolaryngology, St Mary’s Hospital, Praed Street, London W2 1NY, UK.
Tel.: +44 776 458 7025; fax: +44 207 886 1847; e-mail: roland.hettige00@imperial.ac.uk


Objectives:  To assess tracheostomy care and improve standards following the introduction of an ENT-led multidisciplinary tracheostomy ward round service.

Design:  Prospective third cycle audit.

Setting:  Tertiary academic London hospital serving an inner city population of multi-ethnic background (St Mary’s Hospital, Paddington, London).

Participants:  Patients with a tracheostomy discharged from ITU to general wards.

Implemented actions:  • Establishment of an ENT-led Tracheostomy Multidisciplinary Team (TMDT).

• Weekly TMDT ward round to manage patients with a tracheostomy.

• ENT-led educational and training sessions for allied healthcare professionals.

Main outcome measures:  Compliance with local tracheostomy care guidelines (St Mary’s tracheostomy care bundle) and time to tracheostomy tube decannulation.

Results:  Preliminary results of 10 patients show improved compliance with tracheostomy care guidelines, established in 2004, rising to 94%. Average time to decannulation was significantly reduced from 21 to 5 days (P-value = 0.0005, Mann Whitney Wilcoxon Test). The mean total tracheostomy time was reduced from 34 to 24 days although this was not statistically significant (P-value = 0.13, Mann Whitney Wilcoxon Test).

Conclusions:  The introduction of regular ENT-led multidisciplinary input for patients with a tracheostomy significantly improved compliance with nursing care standards. There was also a reduction in the total length of time tracheostomy tubes remain in situ, with time to decannulation significantly reduced.

Tracheostomy is a procedure predicted to become more common as the demand for intensive care services grows. There are, however, substantial concerns regarding the national standard of tracheostomy care currently provided.1 We have recently reported our initial experience attempting to improve tracheostomy care following the introduction and prospective audit of a tracheostomy care bundle (Fig. 1). Although this has shown some benefit, there was still a shortfall in tracheostomy care which warranted further improvement.2

Figure 1.

 Components of the Milne Tracheostomy Care Bundle.

Our aims were to:

  • 1 Improve patient care and expertise by incorporating an ENT led MDT approach for tracheostomy-related problems.
  • 2 Reduce tracheostomy decannulation time following discharge from ITU.
  • 3 Reduce the number of tracheostomy-related complications.
  • 4 Promote national improvement in tracheostomy care on general wards.

In line with Lewis et al., we used ‘time to decannulation’ as an objective measure of tracheostomy care in addition to compliance rates with the tracheostomy care bundle (Fig. 1).

Materials and methods

The changes implemented included:

  • 1 Introduction of a weekly Tracheostomy Multidisciplinary Team (TMDT) ward round. Team members included an ENT SpR and ST2, speech and language therapist, respiratory physiotherapist and critical care outreach nurse. All tracheostomy patients were referred to the respiratory physiotherapist at least 24 h prior to ITU discharge. Emergency referrals were made directly to the ENT team. The TMDT reviewed most patients within 24 h of ITU discharge.
  • 2 Monthly teaching sessions organised for nursing staff involved with tracheostomy care. Material covered was identified from tracheostomy-related clinical incident logs including practical aspects of tracheostomy care, case-based discussions and warning signs of impending problems.
  • 3 ENT-led training day for physiotherapists and speech and language therapists. The emphasis was on clinical skills, tracheostomy inner tube care, tube change and decannulation.

Data collection

A total of 148 patients underwent tracheostomy in the trust in the 18 months period prior to initiation of the TMDT ward-round. Of these 148, 69 patients (47%) either died, were transferred to another trust, or did not have the recorded information regarding their tracheostomy status. This left a cohort of 79 pts which was retrospectively analysed (Table 1). This was compared to data prospectively collected on patients reviewed by the TMDT over a 3-month period (Table 2).

Table 1.   Cohort of patients seen prior to implementation of ward round
Method of insertion and specialtynMean ageMean APACHE II scoreMean LOS on ITU Indication* emergency :  respiratory weanMean total tracheostomy time (days)Time to decannulation post discharge from ITU
  1. *Indication for tracheostomy-emergency (unstable or obstructed airway): respiratory wean.

  2. LOS, length of stay.

Percutaneous4360.82118.02 : 4134.923.0
Surgical-other2863.323203 : 2534.321.1
Surgical-ENT871.62222.32 : 6 27.410.6
Overall79632119.06 : 7333.921.1
Table 2.   Cohort of patients seen on ward round
Method of insertionnMean ageMean APACHE II scoreMean LOS on ITU Indication* emergency :  respiratory weanMean total tracheostomy time (days)Mean time to decannulation post discharge from ITU
  1. For number of times each pt was seen on ward-round please consult Appendix 1.

  2. *Indication for tracheostomy-emergency (unstable or obstructed airway) : respiratory wean.

  3. LOS, length of stay.

Percutaneous44323290 : 424.34.0
Surgical-ENT66223271 : 523.56.3
Overall10542327.31 : 923.85.4

There was no significant difference between the two cohorts regarding age, APACHE II scores, length of ITU stay, type and indication for tracheostomy (Fisher’s Exact Test).

Compliance to the tracheostomy care bundle was recorded for the cohort of patients seen by the TMDT ward-round and compared to the compliance data from a previous prospective second cycle audit cycle performed between November 2006 and February 20072 (see Table 3).

Table 3.   Compliance data for cohort 2 versus second audit cycle2
Components2006–2007 audit, = 70 (% compliance)COHORT 2 = 10 (% compliance)
Adequate humidification66/70 (94)9/10 (90)
Inner tube care documented27/70 (39)10/10 (100)
Dressing changes documented6/70 (9)10/10 (100)
Safety equipment67/70 (96)10/10 (100)
Documentation of cuff status38/70 (54)10/10 (100)
Weaning plan44/70 (63)7/10 (70)
Care plan35/70 (50)10/10 (100)
Overall compliance283/490 (58)66/70 (94)


Between January 2006 and July 2007, 79 patients underwent tracheostomy (Table 1). Tracheostomy performed and managed by the ENT team had a shorter mean decannulation time following discharge from ITU (= 0.01, Mann Whitney Wilcoxon Test).

Between July and September 2007, 10 patients who underwent tracheostomy were reviewed by the TMDT (Table 2).

The mean time to decannulation following ITU discharge was significantly reduced from 21 to 5 days (P-value = 0.0005, Mann Whitney Wilcoxon Test). The total tracheostomy time was also reduced from 34 to 24 days, although this difference was not statistically significant (P-value = 0.13, Mann Whitney Wilcoxon Test).


Synopsis of key findings

Prior to establishment of the TMDT, tracheostomy insertion and management performed by the ENT team had a shorter mean decannulation time following discharge from ITU compared to other specialities (= 0.0136, Mann Whitney Wilcoxon Test). The average decannulation time following discharge from ITU was 21 days. Compliance with the tracheostomy care bundle was 58%.

Following implementation of the TMDT ward-round, nurse education sessions and physiotherapist training days, the pilot cohort of 10 patients had a significantly reduced mean decannulation time of 5 days. Both groups were matched for age, length of stay on ITU, APACHE scoring, type and indications for tracheostomy. Compliance with the care bundle increased to 94% from 58%. Hundred percentage compliance was not achieved because three patients were decannulated before a weaning plan was made and one patient did not receive humidified oxygen (Table 3).

Sixty-nine of 148 patients were not analysed in the pre-TMDT ward-round group, which may introduce an element of selection bias into cohort 1. Also, in light of the small sample size in cohort 2, one must be cautious in drawing firm conclusions from this data, but these preliminary results of our ongoing third cycle audit are encouraging.

Comparisons with other studies

There is national concern regarding tracheostomy care. Of 103, NHS trusts providing a tracheostomy service in the UK, 14.6% (= 15) were found to have no tracheostomy care policy in place, 31% had a dedicated tracheostomy care team and only 12.5% of Trusts had ENT input with tracheostomy care.1

Although the implementation of the TMDT did not significantly reduce the total tracheostomy time, we did significantly reduce the time from ITU discharge to tracheostomy decanulation. This has been used by others as an objective measure of tracheostomy care. Leung et al. postulated that the only predictors for early decannulation are indication for tracheostomy and patient diagnosis.3 In an attempt to reduce bias regarding these two variables, our comparison groups were matched for ‘Indications for Tracheostomy’ and APACHE-II scores. The reduction in decannulation time seen in our pilot cohort is therefore likely to be attributable to improved tracheostomy aftercare.

Norwood et al. published a study with similar interventions to our own carried out in an ITU setting.4 He concluded that fewer tracheostomy patients on general wards, a tracheostomy outreach service and better nurse education resulted in significantly fewer tracheostomy-related complications. This corroborates our view that a reduction in tracheostomy decannulation time on general wards equates to a shorter interval in which tracheostomy – related complications may arise (Appendix 2).

The TMDT provides a beneficial skills mix for optimal patient management. ENT input enables bedside flexible laryngoscopy/tracheoscopy assessment which can facilitate clinical decisions in certain situations. An example in our pilot cohort was a patient with new onset haemoptysis, assumed to be related to granulation tissue associated with the tracheostomy tube. Early identification on the MDT ward round and subsequent ENT assessment revealed the true source of bleeding, arising from the left main bronchus rather than the tracheostome.

Mace et al. corroborate that ENT involvement improves tracheostomy service provision as they provide unique skills in laryngeal/tracheal assessment that other healthcare professionals may lack.1 Law et al. suggested that prior to decannulation, all patients should undergo ENT examination of the airway to detect potential tracheal abnormalities.5

Clinical applicability of the study

The implementation of an ENT-led TMDT, with a weekly ward round and allied healthcare profession teaching programme has resulted in significant improvements with compliance to the tracheostomy care bundle. Mean decannulation time following ITU discharge was also reduced in our pilot cohort. Improvements in tracheostomy care are likely to reduce the number of tracheostomy-related complications encountered on general wards because they can be prevented or treated at an early stage. The TMDT and other implemented changes are set to continue which will enable data collection from a larger cohort of patients. We hope that our preliminary findings will encourage other NHS trusts to assess tracheostomy service provision and initiate similar improvements to benefit patients throughout the NHS.

Conflict of interest

None to declare.