Audit of double-lumen endobronchial intubation

A. H. Seymour*, B. Prasad and R. J. McKenzie

Department of Anaesthetics, Birmingham Heartlands Hospital, Bordesley Green East, Birmingham B9 5SS, UK

* Corresponding author. E-mail: alan.seymour{at}heartsol.wmids.nhs.uk

Accepted for publication May 31, 2004.


    Abstract
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Background. Many technical variations are possible in the placing and management of a double-lumen tube (DLT). We surveyed our practice to relate these variations to the course of the anaesthetic.

Methods. We used a questionnaire to obtain details of technique in 506 consecutive double lumen intubations. The details were related to the incidence of secretions, tube displacement, and decreases of oxygen saturation (<88%) during one lung anaesthesia (OLA).

Results. Robertshaw tubes were used for 482 of the 506 intubations. During OLA there were 48 instances of desaturation (<88%), 19 cases of upper lobe obstruction, 15 of carinal obstruction, 16 of isolation failure, eight of excessive secretions (none of whom had received an antisialogogue; P<0.01) and 12 miscellaneous events. The experience of the anaesthetist or use of a fibre-optic bronchoscope did not affect these events. Air was of no advantage as a maintenance gas. Atropine 400–600 µg appeared to prevent desaturation on OLA (P<0.05) but glycopyrrolate 200 µg did not.

Conclusion. Most factors had little effect on the progress of the anaesthetic, but an antimuscarinic usefully reduced secretions, and atropine (but not glycopyrrolate) was associated with less desaturation during OLA.

Keywords: anaesthetic techniques, regional, one lung ; anaesthetic techniques, regional, thoracic ; anaesthetics gases, nitrous oxide ; antagonists muscarinic, glycopyrrolate ; equipment, double lumen tubes ; parasympathetic nervous system, atropine


    Introduction
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Many variations are possible in the use of a double lumen endobronchial tube. We conducted a prospective survey in our department to determine if these influence the course of the anaesthetic.


    Methods
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
The survey was an audit and did not require Ethical Committee approval. No identifiable patient data were stored in electronic format. We set out to review 500 consecutive double-lumen intubations carried out at Birmingham Heartlands Hospital over a period of 6 months. Anaesthetists were asked to complete a questionnaire each time a double-lumen tube (DLT) was used. The forms were collected regularly and related to the operation register. Where no form was present for an appropriate register entry, a blank form with a survey number was sent to the anaesthetist for completion, and a record kept to ensure that data retrieval was complete. The information sought was:

  1. The experience of the anaesthetist (total thoracic operating sessions done).
  2. Basic patient information: age, height, weight, operation.
  3. The type and size of tube.
  4. Whether the tube was placed with the aid of a fibre-optic bronchoscope (FOB).
  5. Whether an antimuscarinic was used: agent, dose, and time of administration.
  6. The method of cuff inflation.
  7. The method of tube fixation.
  8. The technique used to maintain anaesthesia.
  9. Whether or not excessive secretions were present.
  10. Any problems with the position of the tube that occurred during surgery.
  11. Any episode, however brief, in which saturation levels fell below 88% during one lung anaesthesia (OLA).

Further questions related specifically to the performance of the tube but these results require separate consideration. Each form was evaluated by a senior thoracic anaesthetist (A.S.) to ensure the most accurate possible interpretation of the information. If necessary the anaesthetist concerned with the case was consulted.

All the tubes were initially placed using inspection, listening for leaks, assessing the relative compliance of the two lungs, and auscultation. Reference to a pressure/volume loop trace was sometimes helpful. Operators using the FOB would then confirm the position visually and make any minor adjustments. The system used was to identify the shoulder of the bronchial cuff, at or just beyond the main carina, via the tracheal orifice. Then the bronchial side would be inspected to ensure, on the left, that the orifice of the upper lobe was not obstructed and, on the right, that the orifice of the right upper lobe could be seen through the slit in the cuff. Having turned the patient into the lateral position, most anaesthetists accepted a clinical position check. Two anaesthetists routinely checked the tube position again with the FOB. Subsequent intraoperative checks with the FOB were unusual, and diagnosis of tube displacement was primarily clinical. In about half of the cases of suspected tube displacement, this was followed by inspection by FOB. In the other instances, increased lung compliance and increased arterial oxygen saturation after re-positioning were accepted to indicate that the tube had been displaced. The results were analysed by spreadsheet (Excel 97, Microsoft Corporation, Seattle, WA, USA). Some patients developed more than one problem: if these were related (for example right upper lobe obstruction followed by failure of lung isolation in the case of a short right main bronchus) the event was only entered once for statistical analysis. The {chi}2-test, with Yate's correction when appropriate, was used to test for statistical significance, with the help of SISA at http://home.clara.net/sisa/twoby2.htm (first accessed in August 2003).


    Results
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Patient details are given in Table 1. No missing data were detected during the survey: this is important because we noted that failure to complete questionnaires at the time could occur in emergencies and if the anaesthetist had problems with management of the tube, which was clearly highly relevant to the survey. The most frequent adverse events were tube displacement (or unrecognized misplacement), desaturation during OLA and excessive secretions.


View this table:
[in this window]
[in a new window]
 
Table 1 Patient details

 
Out of the 506 reports, 442 cases were anaesthetized by 12 consultant anaesthetists who regularly anaesthetized such cases, and 25 (mainly emergencies) by other consultants. Specialist registrars managed 39 cases without immediate supervision. A sequential check indicated that the rate of reporting problems was not affected by the fact that the audit was in progress.

Tubes used
For the 506 intubations, a Robertshaw tube (Phoenix Medical Ltd, Preston, UK) was used in 482 (Table 2). For the other 24 cases, a Bronchocath (Mallinckrodt Ltd) was used, and in 14 of these, this was because a Robertshaw tube could not be successfully placed. In only 18 instances, an ipsilateral tube was deployed—usually on the left. The small number of Bronchocath tubes was too small to affect the statistical analysis.


View this table:
[in this window]
[in a new window]
 
Table 2 Distribution of Robertshaw tubes used by side and size

 
Problems during surgery
In 91 of the 506 patients (18%) 119 problems occurred during the surgery. There were 48 instances of desaturation during OLA, 19 cases of upper lobe obstruction (11 left, eight right), 16 of carinal obstruction and 16 instances of right to left or left to right isolation failure. There were eight reports of excessive secretions. In 12 patients, other complications were noted. These were three cases of supraventricular arrhythmia, rupture of one tracheal and two bronchial cuffs, two air leaks (one parenchymal and one via a misplaced suction catheter), two failures of the lung to deflate during thoracoscopy (probably caused by residual subatmospheric intrapleural pressure on the side of the procedure), one case of bronchospasm and one cardiac arrest associated with hypovolaemia. Problems were more frequent with right-sided procedures (19.8%) than with left (15.3%) but this difference was not significant.

Experience of anaesthetist
In no case did a trainee working alone seek senior assistance and there was no relationship between the experience of the anaesthetist and the incidence of complications during anaesthesia.

Checking tube position
A fibre-optic bronchoscope was used to check 283 (56%) of the 506 tube placements. This did not affect the incidence of problems (19%) compared with 17% in those in whom the FOB was not used.

Method of tube fixing
Tubes were either tied or strapped into place, and a Guedel airway was used in some cases next to the tube. These factors did not appear to influence the frequency of DLT displacement, which was close to the mean of 9% with three of the four possible combinations. However, the one anaesthetist who used a firm tie and a Guedel airway reported a displacement rate of 2% in 56 cases (P=0.1).

Endobronchial cuff inflation
A variety of methods were used. Some individuals find it helpful to inflate the cuffs themselves rather than asking an assistant but this had no measurable effect either upon primary location problems or those developing during surgery. Where nitrous oxide was used for anaesthesia the choices included inflating the cuffs with air, saline, or a gas mixture including nitrous oxide. If nitrous oxide was not used for anaesthesia, air or saline was used for cuff inflation. None of these combinations affected the rate of tube displacement. There were three cuff ruptures. One bronchial cuff failure was caused by a surgical needle. In the other two (one tracheal and one bronchial) both the cuffs were inflated with air and the maintenance included nitrous oxide (P<0.05). The burst tracheal cuff was on a right-sided 41fr. Bronchocath.

The effect of an antimuscarinic agent
Secretions can impair gas exchange, especially in patients with diseased lungs. In a DLT, with long narrow internal dimensions, this could be worse. There were only seven reports of excessive secretions. All were in patients that had not received a previous antisialogogue (P<0.01). Four of these patients had episodes of oxygen desaturation during OLA. We therefore studied if an antimuscarinic agent, given at induction, affected saturation during OLA. We did not consider cases where upper lobe obstruction was a contributory factor, and also did not consider a patient from ITU. Of the 497 considered, 274 received an antimuscarinic and 223 did not. In the patients given atropine or glycopyrrolate, 16 (6%) showed desaturation during OLA compared with 23 (10%) of those who did not (P=0.07). Considering those given the antimuscarinic, 116 were given atropine in doses between 200 µg (in one instance), up to 600 µg, whilst 158 were given 200 µg of glycopyrrolate. The effect of glycopyrrolate compared with no medication was insignificant: 12 (7.5%) showed desaturation compared with the 23 (10%) (P=0.25). When atropine was given only four patients (3.5%) showed desaturation (P<0.05).

Influence of maintenance regime
A vapour/air/oxygen mixture was used for 252 patients and vapour/nitrous oxide was used for 249. This did not affect the outcome.


    Discussion
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
We confirmed that antimuscarinics reduce the secretions of patients undergoing OLA. We also found, unexpectedly, that atropine, but not glycopyrrolate, appears to reduce desaturation during OLA. Glycopyrrolate had a statistically insignificant benefit in respect of desaturation. Whilst most patients were given 600 µg of atropine, 32 (27%) had amounts varying between 200 and 400 µg, which seemed to work just as well. The following ways that oxygenation may be improved are possible. First, atropine may be more active in reducing secretions, which seem to lead to desaturation. Another possibility is that atropine may have a more favourable effect than glycopyrrolate upon alveolar gas mixing,1 and hence alveolar ventilation, but no comparative study is available. Finally, confounding factors could be present including ventilation methods, , or the application of positive end-expired pressure to the ventilated lung. We were unable to find any such factors. Atropine therefore seems to be advisable. A reason for avoiding atropine is its effect on heart rhythm. We did not specifically seek information about cardiac arrhythmia but in the instances where it was reported atropine had not been used.

We found three important negative features. Checking tube position with the FOB seemed to confer no advantage, even accounting for the fact that some operators reserved direct visual placement for more difficult cases. This supports the report by Seymour and Lynch2 in a different group of patients. The experience of the anaesthetist appeared to have no effect on the course of a thoracic anaesthetic, but this could have been affected by case selection. Lastly, the use of air rather than nitrous oxide in the gas mixture was not beneficial. Nitrous oxide might have been avoided in sick patients, but in practice this did not happen in more than one or two instances.

Finally, we made two equivocal observations. First, the good showing from tying the tube in place, with a Guedel airway as support, warrants a formal trial. Secondly, the ability of nitrous oxide to diffuse into air-filled cuffs, increasing their pressure or volume, is well known.3 If it were possible to be sure that the cuffs that ruptured when this was done were not compromised in some way, the finding would be significant, but the number of cases is not sufficient to be sure. It is perhaps surprising that the rate of tube displacement in the 87 instances where the combination occurred was just the same as for all the other groups.

In conclusion, most aspects of thoracic anaesthesia are little affected by the techniques used with the DLT. Patients benefit from an antimuscarinic to reduce secretions before double-lumen intubation. In the doses used, atropine, but not glycopyrrolate, reduced desaturation during OLA.


    References
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
1 Kox W, Langley F, Horsfield K, et al. Effect of atropine on alveolar gas mixing in man. Clin Sci (Lond) 1982; 62: 549–51[Medline]

2 Seymour AH, Lynch L. An audit of Robertshaw double lumen tube placement using the fibreoptic bronchoscope. Br J Anaesth 2002; 89: 661–2[Free Full Text]

3 Karasawa F, Ohshima T, Takamatsu I, et al. The effect on intracuff pressure of various nitrous oxide concentrations used for inflating an endotracheal tube cuff. Anesth Analg 2000; 91: 708–13[Abstract/Free Full Text]