We thank Dr Magee, who is absolutely correct in pointing out the fundamental error in our description [1] of the findings of Rosen and Hillard [2]. In the editorial we intended to emphasise that, as Rosen and Hillard demonstrated, a potential penalty of moving to smaller tubes is that the ratio of the internal diameter of these tubes to that of large suction catheters or bronchoscopes is more likely to increase the likelihood of generating negative intrathoracic pressure.

In his letter, Dr Magee also elaborates eloquently on other aspects of cuff design that impact on tracheal wall pressure but we are sure he would agree that sore throat is a very general effect from a multiplicity of causes, of which tracheal wall pressure is one.

We were delighted to read the brief but succinct description of the impact of turbulent flow on the work of breathing and the relationship of flow to the square root of the pressure gradient, as a general approximation. Clearly the work of breathing is increased with turbulent flow and it seems very likely to us that in the trachea during normal, if not exertional, breathing, one might expect turbulent flow for at least part of the respiratory cycle, if not most of it.

We would also like to thank Mir and colleagues for their letter. It is always useful to see the problem from the perspective of another speciality, especially one that deals with the complications of our actions. The relevance of the tube's length is important and the benefits of tracheostomy clear. We are not advocating long-term intubation with tracheal tubes but issues frequently arise in the first few days of intubation, before tracheostomy. We are also aware that tracheostomies are not innocuous either, in terms of immediate or late problems including airway narrowing. We agree that selection on the basis of male/female alone would be absurd, but it is a reasonable rule of thumb to which most anaesthetists then apply common sense.

Mir et al. will be aware of the significant practicalities of managing day-to-day intubation in an ICU during critical illness, including the frequent and potentially life-threatening issues of suctioning viscid secretions and bronchoscopy with a narrow tube in place and the risks inherent in tube blockage or exchange. Not every patient with a tracheal tube needs a tracheostomy, but weaning after several days’ intubation with a relatively narrow tube is a problem in some patients. It is possible that failure to wean with a 7.5-mm tube is in itself an indication for tracheostomy, but that has yet to be investigated.

The study quoted by Mir et al. is very interesting and useful [3]. With 1175 tracheostomies it is a large and retrospective tracheostomy study, not a tracheal tube study. All the patients with tracheal tubes subsequently had tracheostomies, with 20 cases of airway stenosis as a late complication of tracheostomy (1.7%) and 104 stenosis cases in total including the follow-up period (0.04% in non-obese and 9.6% in obese). In these 104 patients the tracheostomy size, technique and difficulty of insertion, and total duration of intubation/tracheostomy cannulation is unclear. There was also an 18% re-intubation rate; the association between this and stenosis is unclear. In the obese patients there is no mention of patients’ height, other than BMI, so were the larger tubes in the larger obese patients, did they stay in longer, or did they have other issues such as difficult tracheostomy?

The reported rates of stenosis after tracheostomy are very variable (<0.5–31%) depending on how they were assessed and whether there were clinical implications [4-7]. There is no comment about the relative roles of tracheal intubation and tracheostomy in causing airways stenosis in Halum et al.'s study, although one might assume that the posterior glottic stenoses are more likely to be tracheal tube-related, while the rest could be either. We note that information regarding decannulation was available in 40.1% of all patients, with 39.6% of these undergoing decannulation during the mean follow-up period of 25.3 weeks. Associations in retrospective studies are always interesting and should always raise questions, but should rarely be considered proof of cause and effect. The National Centre for Airway Reconstruction database should surely have information on both duration of intubation and tube size, as well as the precise location of the stenosis and its possible mechanism, i.e. cuff or tube. It would be interesting to see those figures.

The mechanisms that cause tracheal injury and subsequent stenosis are many and variable, and while tube size may be associated, so is tracheostomy. As with all ICU procedures there is a risk-benefit balance; we accept there may be problems with larger tubes but there are also realistic benefits in critically ill patients versus immediate hazards imposed by narrower tubes. A 7.5-mm tube makes both suctioning and bronchoscopy more of an issue and any encrustation will have a significantly greater impact on air flow. Our position remains that, in the critically ill, the rule-of-thumb use of 8- and 9-mm tracheal tubes in adults improves the ease of management and the margin of safety, although we are aware there may be a small but significant complication rate, as with most of our procedures.


  1. Top of page
  2. References