A Comparison of the Reliability of Two Techniques of Left Double-Lumen Tube Bronchial Cuff Inflation in Producing Water-Tight Seal of the Left Mainstem Bronchus

A double-lumen endobronchial tube (DLT) bronchial cuff inflation technique that reliably ensures effective water-tight isolation of the two lungs has not been determined. In this study, 20 patients undergoing thoracic surgery requiring a left DLT had the bronchial cuff of the DLT inflated by one of two techniques. In Group 1, the cuff was inflated to produce an air-tight seal of the left bronchus using the underwater seal technique. In Group 2, the cuff was inflated to a pressure of 25 cm H2 O. After bronchial cuff inflation in both groups, water-tight bronchial seal was tested by instilling 2 mL of 0.01% methylene blue (MB) above the bronchial cuff of the DLT. Fifteen minutes later, fiberoptic bronchoscopy was performed via the bronchial lumen of the DLT to determine whether MB had seeped past the bronchial cuff. Cuff volume was 0.75 +/- 0.64 and 0.76 +/- 0.46 mL, cuff pressure was 30.1 +/- 27.0 and 25.0 +/- 0.0 cm H2 O (mean +/- SD), and MB was positively identified in two and five patients in Groups 1 and 2, respectively. The difference in cuff volume and pressure and the higher MB seepage in Group 2 compared with Group 1 was not statistically significant. In both groups, MB seepage occurred only when the bronchial cuff volume was <1 mL and when the patients were positioned in the left lateral decubitus position. These findings suggest that the risk of aspiration is greatest when the DLT is positioned in the dependent lung and when the bronchial cuff volume is <1 mL. Implications: Water-tight sealing of the left bronchus by DLT bronchial cuff was tested after cuff inflation using two different techniques. Neither air-tight bronchial seal nor cuff pressure of 25 cm H2 O guaranteed protection against aspiration. The risk of aspiration was greatest when the DLT was positioned in the dependent lung and when the bronchial cuff volume was <1 mL. (Anesth Analg 1998;87:1027-31)