Letters

Recent advances in intensive care

Percutaneous tracheostomy may not be more effective than open technique

Author's reply

More still needs to be known about immunonutrition

Percutaneous tracheostomy may not be more effective than open technique

EDITORIn his review of recent advances in intensive care Stott emphasised the efficacy of percutaneous tracheostomy over the conventional open surgical approach.¹ We agree that percutaneous tracheostomy may indeed be the preferred method in selected cases, but it is invariably not used in patients who present anatomical challenges such as a short, thick neck; goitre; a history of neck surgery; or a concurrent coagulopathy. Such difficult cases probably represent about a quarter of an average population requiring tracheostomy.

We disagree with Stott's assertion that the incidence of complications has been shown to be lower with percutaneous tracheostomy than with open surgical techniques. The paper quoted in support of this statementby Hill et aldescribes the results of 356 percutaneous procedures performed in a single unit over four years.² No direct comparison is made with open tracheostomies, and the authors do not report the numbers and type of patients deemed unsuitable for percutaneous tracheostomy and referred for open procedures. The six previous independent series quoted that detailed complication rates for open tracheostomy were published on average nearly 20 years earlier, and a conclusion was reached by comparison with these older studies.

Stott fails to mention Dulguerov et al's meta-analysis reported in 1999, which showed that percutaneous tracheostomy is associated with a higher prevalence of certain complications.³ Their comparison of recent surgical tracheostomy trials (n=21; 3512 patients) and percutaneous tracheostomy trials (n=27; 1817 patients) shows that perioperative complications are more common with the percutaneous technique (10% v 3%), whereas postoperative complications occur more often with the surgical technique (10% v 7%). The bulk of the differences is in minor complications, except perioperative death (0.44% v 0.03%) and serious cardiorespiratory events (0.33% v 0.06%), which were much higher with the percutaneous technique. These authors also noted significantly fewer complications in recent studies compared with older ones.

In the present state of knowledge it would seem unethical to mount a large double blind trial comparing the two techniques as this would subject about a quarter of patients, who would usually be deemed unsuitable for a percutaneous technique, to an increased risk of complications. It is likely that such patients will continue to be referred for open tracheostomy. While this selection bias continues, it is misguided to state that percutaneous tracheostomy holds fewer risks than the open technique.

Natalie Brookes, specialist registrar. 
David Howard, senior lecturer. 
Royal National Throat, Nose and Ear Hospital, London WC1X 8DA

Author's reply

EDITORBrookes and Howard's point that not all patients are suitable for a percutaneous tracheostomy is well made, but the figure quoted of 25% is conjecture and not referenced. As I cannot find any data to produce a figure of my own I will not, but I do question their final paragraph when they state that because of this figure a double blind trial would be unethical. This is curious as nowhere in my article did I suggest that one should be performed.

The meta-analysis referenced looks only at studies published up to 1996 and compares four different percutaneous techniques with surgical tracheostomies performed in two different eras.¹ As most of the deaths occurring with percutaneous tracheostomy occur with a non-dilatational technique it would seem prudent to exclude these from the analysis.

There have been three published prospective trials comparing dilatational percutaneous tracheostomy with open surgical techniques.^2-4 These show complication rates for percutaneous tracheostomy to be as low as or lower than those for the surgical technique. This, coupled with the reduced cost, avoidance of moving critically ill patients, and low long term complication rates, means that I can still conclude that percutaneous tracheostomy is a significant recent advance.

Stephen Stott, consultant in anaesthesia and intensive care. 
Grampian University Hospitals Trust, Aberdeen AB25 2ZN s.a.stott{at}abdn.ac.uk

More still needs to be known about immunonutrition

EDITORI have concerns over one section in Stott's article on recent advances in intensive care.¹ These concerns relate to the paragraphs on nutrition, and particularly the key points box on page 358; there it is stated that recent work has shown that omega 3 polyunsaturated fatty acids improve mortality in critical illness.

The question of immunonutrition is complex, and the different nutrients used in clinical trials can have different, and opposing, effects on the immune system. For example, L-arginine and glutamine stimulate a variety of immune functions.² In contrast, omega 3 and omega 6 fatty acids inhibit a variety of immune functions. ^{2 3}

The timing of administration of these nutrients may therefore be critically important: although it is generally believed that stimulation of the immune system is beneficial in critically ill patients, this may not always be the case. For example, in certain patients (those with adult respiratory distress syndrome or multiorgan failure) in whom the cytokine cascade and production of inflammatory mediators has been suggested to be excessive, administration of omega 3 and omega 6 fatty acids would be beneficial in reducing production of these mediators.

In terms of immunonutrition affecting clinical outcome in critical illness, a recent critical analysis of all the randomised controlled trials in which L-glutamine was given found little evidence to support its routine clinical use.⁴ A recent meta-analysis found that giving combinations of immunomodulatory nutrients reduced infectious complications but not mortality.⁵

When the use of nutrition in critically ill patients is being considered, these considerations must be taken into account. It is still not clear which patients will benefit from being fed these immunomodulatory nutrients and which may not.

Steven D Heys, professor of surgical oncology and nutritional oncology. 
University of Aberdeen, Aberdeen AB24 3UX s.d.heys{at}abdn.ac.uk