A common question asked in the anaesthetic room is “what size of tube?”. Ultimately, the intended purpose and duration of intubation should govern the choice of size. In 1928, Magill suggested “the largest endotracheal tube which the larynx will comfortably accommodate” . Even 30 years ago, it was common practice to place 9.0- or even 10.0-mm tubes for males and 8.0-mm tubes for females. Over recent years it has become common for anaesthetists to place tubes 1–2 mm smaller. This evolution in practice was initially driven by the observation that there was little impact on ventilator pressures during anaesthesia, and is maintained by a perceived reduction in sore throat and hoarseness, increased ease of insertion and less tracheal damage.
The incidence of postoperative sore throat after tracheal intubation varies from 14% to 50%, as does hoarseness, although the latter is rarely sustained [2–7]. Contributory factors include the size of the tube, cuff design and pressure, variation in skills and techniques between anaesthetists and the subjectivity of the symptom of sore throat in individual patients [5, 8–14]. It has been clearly shown that smaller tubes cause less sore throat, with the incidence halved when tubes were reduced from 9.0 mm to 7.0 mm for men and from 8.5 mm to 6.5 mm for women, although these are short-term effects and the incidence of occasional prolonged symptoms seems independent of tube size [5–7]. Be that as it may, sore throat is not limited to patients subjected to tracheal intubation but is also seen with supraglottic airways; even with the laryngeal mask airway (LMA) variable numbers of patients (14–42%) may complain of sore throat postoperatively, a figure not dissimilar from – but usually lower than – that following tracheal intubation [4, 15, 16]. Larger LMAs are associated with a higher incidence of sore throat and hoarseness .
Cuff design is important as with the old red rubber tubes, the cuff pressure could readily contribute to mucosal ischaemia or inflammation . Even now, failure to monitor cuff pressure can result in high pressures . Thin-walled, low-pressure, high-volume cuffs should preclude this problem although it is suggested that the high-volume cuffs are associated with a higher incidence of sore throat due to the greater surface area of cuff-tracheal contact. This is despite the fact that the actual mucosal damage is less than with low-volume cuffs [10, 20–24]. Paradoxically, Loeser et al. showed that uncuffed tubes were also associated with sore throat, so the cuff is not the entire problem .
There can be little argument with Stenqvist et al.’s demonstration of minimal changes in ventilator pressures during routine anaesthesia using tubes as small as 6.0 mm . Spontaneous breathing through a tracheal tube during anaesthesia is a largely redundant technique with the advent of the LMA and most healthy patients can cope with a few minutes’ breathing through a small tube at the end of anaesthesia. Smaller tubes may be easier to insert, including using fibreoptic endoscopy, as the view of the larynx during passage of the tube is subjectively better . Insertion is likely to be less traumatic and there is an association between larger tubes and glottic and tracheal damage, particularly in women [27, 28]. Computed tomographic imaging six months after intubation has demonstrated some degree of laryngeal abnormality including tears, scars and laryngocoeles in 86 of 100 patients, although the clinical implications of these findings are unclear . For anaesthesia, these arguments all favour the use of a smaller tube, hence the logical move in this direction.
In the intensive care unit (ICU) population, however, the purpose and duration of intubation is often very different compared with that during anaesthesia.
In the ICU, patients may require tracheal intubation for a longer period of time. The function of the tube is not only a conduit for ventilation but also the point of access for airway toilet to clear secretions, and to allow safe passage of a fibreoptic bronchoscope for diagnostic and therapeutic purpose. Bronchial toilet is a common and important requirement in the ICU. Bulky secretions require substantial suction catheters for efficacy; with smaller tubes, the catheter may obstruct a significant portion of the tube and pressure equalisation is dependent on ‘air’ entrainment between the catheter and the tube. A 4.0-mm catheter will occlude a significant part of a 6.0- or 7.0-mm tracheal tube as well as being susceptible to blockage by thick or bulky secretions. Rosen and Hillard found that if the ratio of the outer diameter of the suction catheter to the internal diameter of the tracheal tube exceeded 0.5, no large negative pressure will be applied across the intrathoracic space, but if it is less than this the resultant negative intrathoracic pressure can result in atelectasis and cardiovascular compromise . This problem is potentially increased by bronchoscopy: the standard adult ICU fibreoptic bronchoscope has a diameter of 5.7 mm with a 2-mm suction channel to enable adequate suction. This limits the tracheal tube to those larger than 7.5–8.0 mm. With smaller tubes, the bronchoscope can be difficult to insert and may even become irretrievably lodged within the tube. Even with a 8.0-mm tube, the bronchoscope occupies more than half of the effective tube diameter (Table 1), which can lead to increased airway pressures, increased auto-PEEP, reduced tidal volumes, hypoxia and hypercarbia . Furthermore, driving pressures as high as 70 cmH2O can be produced, with a reduction in tidal volume by as much as 80% of the set value .
|6.0 mm*||7.0 mm||8.0 mm||9.0 mm|
|Cross-sectional area of tube without bronchoscope; mm2||28.3||38.5||50.3||63.6|
|Remaining tube area with bronchoscope in situ; mm2||6.8||17.0||28.7||42.1|
|Proportion of tube cross-section area obstructed||76.0%||55.8%||42.9%||33.8%|
Patients in the ICU often have compromised respiratory function and will need to be weaned from ventilatory support, and hence may be expected to breathe through the tracheal tube over long periods of time. Work of breathing can be defined in its most basic form as a product of pressure and volume. Shapiro et al. found that in normal volunteers breathing through tracheal tubes (6.0–9.0 mm) at a constant tidal volume of 500 ml, work of breathing and tension-time index increased as tube diameter decreased . The work of breathing did not become excessive until minute ventilation exceeded 15 l.min−1; however, patients in the ICU are far more likely to be clinically compromised by relatively small changes . Bersten et al. used a lung simulator to show that work of breathing ranged from 0.009 kg.m.l−1 with 9.0-mm tubes to 0.25 kg.m.l−1 with 6.0-mm tubes, the latter representing 490% of the normal work of breathing . Work will be determined in part by the pressure gradients generated for flow to occur. The relationship between these gradients and tube diameter is well known from the equation described by Hagen-Poiseuille for laminar flow: if a tube’s radius is reduced by half, the flow through the tube decreases sixteen-fold or the pressure gradient has to increase. Turbulent flow is mathematically complex and no simple formula exists, but the effect is to increase the resistance, and hence pressure changes, further and unpredictably. The transition from laminar to turbulent flow is described by a Reynolds number of around 2000 and the critical flows at which this occurs are shown in Fig. 1. The peak inspiratory flow rate is important; in quiet breathing this may be around 20 l.min−1, but in the vulnerable patient with respiratory compromise, as the respiratory rate rises, values of up to 50–100 l.min−1 are common . This implies that most of the flow will be turbulent already and that the work will further increase with diminishing tube size. It follows that larger diameter tracheal tubes have more reserve before significant increases in airway resistance occur, translating to a better margin of safety.
Another consideration is encrustation of the tube with biofilm, which has the potential to reduce the tube’s diameter even further as well as being implicated in causing ventilator associated pneumonia. Sottile et al. examined 25 tracheal tubes in the ICU with scanning electron microscopy, demonstrating that confluent amorphous material was present on 84% of tubes and intermittent amorphous material was present on all of the remaining 16% . In another study, 38 of 40 tubes examined post-extubation after a minimum of 72 hours’ use, contained debris. The mean overall depth was 0.64 mm, with mean greatest depth 2 mm (range 0–5 mm) . The depth of debris correlates with the duration of intubation . When intraluminal diameter was measured acoustically in 94 tubes in use for a minimum of 12 hours, in almost 60% of patients the volume reduction exceeded 10%, and in 22% of cases this led to a luminal internal diameter <7.0 mm. The reduction to <7.0 mm was less frequent (9.3%) in tubes over 8.0 mm in diameter .
In the critically ill, an important practical consideration is that once a tube is placed, replacing it is often hazardous [38, 39]. The grade of intubation can radically and rapidly change during a stay in the ICU due to underlying pathology, oedema or difficulties with positioning. The bigger problem is illustrated in burns patients where small tubes, consistent with current anaesthetic practice, are often placed in in the emergency department. Changing to a bigger tube may be needed for bronchial toilet for secretions and plugging but may be hazardous because of oedema. While perhaps obvious in a burns patient, these issues are potentially relevant in any postoperative patient, or others going to the ICU for ventilation [40–42]. The anecdotal observation that small tubes are being placed suggests that the concept of considering tube choice in terms of being ‘fit for purpose’ is either overlooked or not known.
Not all the features of larger tubes are positive. Laryngeal injury seen in patients studied at extubation was associated with the height:tracheal tube size ratio but was also associated with emergency intubation and with the duration of intubation. The clinical significance of these visual findings is not clear but most are transient [43, 44].
In conclusion, there are cogent practical reasons for placing larger tubes in critically ill patients while smaller tubes may be perfectly acceptable for routine anaesthesia. We would suggest that if admission to ICU is contemplated then the time-honoured ‘8.0 for females, 9.0 for males’ is a reasonable rule of thumb, unless circumstances dictate otherwise, e.g. in difficult airways or particularly small patients. Whether in anaesthesia or the ICU, the tracheal tube should be ‘fit for purpose’. Times and indications have changed but surely the answer to the question “what size of tube?” might be a modern variation on Magill’s statement: “the most appropriate size for the purpose intended”, bearing in mind that the purpose may extend beyond anaesthesia and into the ICU. More importantly, the same thought and consideration should go into choice of tube as for other more esoteric parts of anaesthetic practice.