There is disagreement among investigations on the most effective method for choosing appropriate double-lumen tube (DLT) size, with some authors recommending the use of various radiological measurements such as left mainstem bronchial diameter, tracheal width, or both.1–7 Other reports recommend the use of complex mathematical formulas for predicting the correct endobronchial tube size.1,5,6 Perhaps, because of the complexity of the latter approach, the choice of left-sided DLT size (35–41 FR) in adults undergoing thoracic surgery requiring one-lung ventilation (OLV) is often based on gender and/or height, with the goal of inserting the largest possible DLT.8–11 With the exception of isolated case reports and observational studies, there are no prospective studies, to our knowledge, evaluating the impact of DLT size on intraoperative outcome measures, particularly arterial oxygen desaturation and lung isolation failure. Hypoxemia during OLV, usually defined as Spo2 <90%, is reported to occur in 1% of patients in one report10 but in 7%–10% of patients in other studies.12,13 Proponents of inserting the “largest tube that will safely fit the airway” suggest that this practice minimizes DLT migration and obstruction of the left upper lobe orifice from a smaller left-sided tube.10,14 The latter practice is proposed to prevent hypoxemia during OLV or proximal displacement, which may result in lung isolation failure and tracheal obstruction. Other physicians choose to insert smaller than conventionally sized DLTs based on a perceived benefit of ease of atraumatic insertion. At our institution, there is a wide variation in practice patterns on DLT size determination, with some physicians routinely choosing smaller than conventional sized DLTs. We, thus conducted a prospective pilot study as part of a quality improvement initiative to evaluate whether the use of smaller DLT than that recommended based on height or gender8–11 is associated with a similar composite incidence of intraoperative hypoxemia, need for DLT repositioning during surgery, or lung isolation failure.
With IRB approval for waiver of informed consent, we prospectively collected data over six mo on 304 consecutive patients undergoing lung or esophageal surgery requiring OLV using a left-sided DLT. Preoperative pulmonary function test results were recorded. Standard anesthesia induction regimens, consisting of propofol 1–3 mg/kg, fentanyl 1–4 μg/kg, vecuronium for muscle relaxation and maintenance of anesthesia using fentanyl and isoflurane in oxygen, were used for all patients.
Endobronchial DLT Insertion and Positioning
We used DLTs from one manufacturer (Sheridan, Telefex Medical, Research Triangle Park, NC). A 28 FR DLT was used in teenagers and very small adults. The Sheridan DLTs have an equal internal diameter for both lumens. The internal diameter and outer diameter of the corresponding DLT are as follows: 28 FR (3.8 and 10.1 mm), 35 FR (4.9 and 13.0 mm), 37 FR (5.2 and 13.6 mm), 39 FR (5.5 and 14.4 mm), 41 FR (5.6 and 14.6 mm), respectively. DLT size was chosen at the discretion of the participating anesthesiologist. Two investigators have a practice of inserting 35 FR DLT for both men and women, regardless of height or body habitus. Two other investigators use a conventional approach of inserting the largest possible DLT according to patient gender and/or height as previously recommended (35 or 37 FR for height <165 cm, 37 or 39 FR for height 165–177 cm, 39 or 41 FR for height >177 cm).11 The DLTs were inserted under direct laryngoscopy and the position then confirmed by fiberoptic brochoscope (FOB), unless preoperative chest radiography demonstrated tracheal deviation requiring positioning with the aid of FOB. A 4.1 mm FOB (Olympus LF-G, Orangeburg, NY) used in our study is compatible for use in 35–41 FR DLT. The bronchial cuff was inflated with <4 mL of air, sufficient to prevent an audible air leak with the operative side isolated.
Management of OLV
During OLV a Fio2 of 1.0 and volume-controlled ventilation were used with a tidal volume of 6–8 mL/kg keeping peak inspiratory pressure <35 cm H2O. The respiratory rate was adjusted to maintain an end-tidal CO2 <45 mm Hg. Arterial oxygen desaturation was managed by first excluding malpositioning or mechanical obstruction of the DLT then by one or more of the following at the discretion of the staff anesthesiologist: intermittent inflation of the dependent lung, positive end-expiratory pressure to the dependent lung, or continuous positive airway pressure to the nondependent lung. Patients were routinely surveyed in the postanesthesia care unit for occurrence of significant airway or vocal cord injuries that might be related to endotracheal intubation. The incidence of hoarseness was not recorded.
The operations were performed with either minimally invasive techniques using video-assisted thoracic surgery or standard thoracotomy designed to remove all neoplastic disease and the ipsilateral mediastinal lymph nodes. Based on surgeon’s preference, postoperative pain relief was provided by continuous administration of either epidural opioid (usually fentanyl and bupivacaine 0.05%) or IV opioid patient-controlled analgesia, usually with morphine. After an overnight stay in the postanesthesia care unit the patients were transferred to the thoracic surgical floor on the first postoperative day.
The intraoperative outcomes recorded on a study data form were 1) the number of attempts required for adjusting the DLT position during surgery; 2) inadequate collapse of the operative lung at any point during surgery; and 3) hypoxemia during OLV, defined as the occurrence of Spo2 ≤88% of any duration. During each case, data were prospectively entered on a designated study questionnaire by residents or attending staff and later recorded by one research assistant. To determine the difference of patient and operative characteristics between DLT size categories, all variables were examined by univariate analysis Student’s t-test or Fisher’s exact test. One-way ANOVA was used to detect differences in patients’ characteristics between the DLT size groups. Data are presented as mean value ± sd unless otherwise indicated and P < 0.05 was considered significant. Statistical analysis was performed with the software SPSS version 10.0 (SPSS, Chicago, IL).
Of the 304 patients studied, 2 receiving a 41 FR and 2 receiving a 28 FR DLT were excluded from the analysis as statistical outliers. Patient characteristics for the study group (n = 300) are shown in Table 1. Data were available from all patients. In males, there were no differences in any characteristic among the DLT groups. Among females, those receiving a 37 FR DLT weighed more than those who had a 35 FR DLT inserted (74 ± 21 vs 66 ± 14 kg, P = 0.03, respectively). In three patients the DLT (two 39 FR and one 35 FR) was believed to be over-sized when no inflation of the bronchial cuff was needed for lung isolation. In six other patients, the initial DLT was replaced with a smaller DLT due to difficulty of insertion into the left mainstem bronchus (one 39 to 37; three 37 to 35; and two 35 to 28 FR). In three patients preoperative chest radiography demonstrated tracheal deviation requiring fiberoptic positioning of the DLT (two 35 and one 37 FR).
The incidence of intraoperative composite outcome measures among the groups is shown in Table 2. Regardless of gender or height, there were no differences in any of the measured outcomes across the three DLT sizes studied. The incidence of transient hypoxemia for all groups was 3.7% (11 of 300). The incidence of lung isolation failure (13 of 300 or 4.3%) and need for DLT repositioning during surgery occurred in 34 of 300 or 11.3% patients. Inadequate lung isolation was usually attributable to malpositioned DLT. There were no observed complications consistent with airway injury related to DLT placement.
The main findings of this study are that, in patients undergoing pulmonary or esophageal surgery requiring OLV with the use of a left-sided DLT, the use of a smaller than conventional DLT size did not influence the composite incidence of intraoperative hypoxemia, need for repositioning of the DLT or success of lung isolation. Our findings were consistent, regardless of gender, height, or DLT size. Because 35 FR DLTs are more frequently used in females in general, our conclusions may have an even greater significance for male patients. Despite the proportionally greater overall use of 35 FR DLTs in this study, 2% of patients required further down-sizing. The limiting factor in these cases was not at the glottis but at the level of the left mainstem bronchus. Using the tracheal diameter to predict DLT size, a study of 66 smaller sized Asian patients found that 29% were considered to have an over-sized DLT and 8% required either down-sizing or abandoning one-lung isolation.3
We were able to show that the use of a 35 FR DLT in all adults was associated with a similar and not worse incidence of the most common clinical outcomes in comparison to conventional sizing recommendations.7,8–10 The incidence of the more devastating complication of airway rupture due to plastic DLT is probably under-reported but, nonetheless, rare.14 Also unknown is the incidence of airway damage short of its rupture due to DLT placement. Intuitively, it may seem beneficial to avoid placing a larger than needed DLT for fear of airway trauma and the possibility of greater vocal cord edema. The phenomenon of postintubation vocal cord edema has been described in several patient populations.15–17 Little data are available on the short- or long-term incidence of hoarseness, vocal cord paralysis, granulomas, arytenoid dislocation and hematomas for example, after endotracheal intubation with a DLT compared with that of standard endotracheal tubes.
Proponents of inserting the “largest tube that will safely fit the airway” suggest that this would prevent adverse outcomes with use of a smaller left-sided DLT such as 1) an increased incidence of hypoxemia due to migration and obstruction of the left upper lobe orifice, 2) need for higher balloon cuff inflation pressures that would result in mucosal ischemia and, finally, 3) greater incidence of airway rupture.10,14 We detected less frequent incidence of hypoxemia (3.7%) than that reported by some (7%–10%)12,13 but not others (1%).10 While the causes for transient hypoxemia during surgery are many, the ready availability of FOB to adjust for the common occurrence of DLT displacement or malposition during surgery18 is considered routine care at our institution.8,9 The overall incidence of temporary inadequate lung isolation, which was usually attributable to a malpositioned DLT, in our study was 4.3% and similar to the occurrence of DLT malpositioning (6.2%) in one observational study.10 Second, it has been recommended by some investigators to keep the bronchial cuff inflation pressure <25 cm H2O to avoid compromise of mucosal blood flow.19 These authors demonstrated that an air-tight seal of the left bronchus by the bronchial cuff does not guarantee a water-tight seal.14 Furthermore, it has been demonstrated that cuff pressures vary by manufacturer of DLT, and that cuff pressure does not necessarily correlate to pressure exerted on the bronchial wall.19 Lastly, tracheobronchial rupture may occur during insertion of styletted DLT beyond the glottis, with over-inflation of the balloon (especially when nitrous oxide is used) but also with small volumes of air and with small or large DLTs.14,19–22 Although it has been demonstrated that leaving the stylet enhances the likelihood that a blindly inserted DLT will be correctly placed,23 this may be responsible for serious airway injuries as is the advancement of the DLT until resistance is met beyond estimated adult depths of insertion.24,25
Our study has several limitations. This was a prospective observational study without formal randomization; however, the study design consisted of a random assignment of cases to two investigators who had one standard of care and two others who used an alternative one. No formal power calculation was done for this pilot quality-improvement project and our intent was to use these data to better estimate the primary outcome rates in a representative population of patients. Although the total sample size was modest (n = 300) and none of the comparisons we considered was significant, there was considerable overlap of confidence intervals. This suggests that it is unlikely that we missed any clinically significant differences (Type II error). We used a slightly more stringent definition for hypoxemia as Spo2 ≤88%, rather than the more commonly used <90%, in order to avoid recording borderline cases. Some might argue that our results may not be applicable to general practitioners, since clinicians with extensive experience in complex thoracic anesthesia were involved. However, the left-sided DLT used was standard and, as stated above, our observed overall rates of hypoxemia or DLT malposition were similar to those reported by others. Finally, our conclusions are limited to a practice of using common sized 35–41 FR DLTs, made by one manufacturer (Sheridan, Teleflex, Research Triangle Park, NC) and no 32 FR DLTs which require a less readily available infant size FOB with limited suctioning capability.8,9 The 4.1-mm FOB used in our study is compatible for use in 35–41 FR DLT.
These findings suggest that the use of smaller than conventional DLTs is associated with similar intraoperative clinical outcomes. We question whether the optimal size of a left-sided DLT is the largest tube whose bronchial lumen fits the desired bronchus. Further studies are needed to define the incidence of vocal cord edema, hoarseness and airway injury related to DLT size.
1. Hannallah MS, Benumof JL, Ruttimann UE. The relationship between left mainstem bronchial diameter and patient size. J Cardiothorac Vasc Anesth 1995;9:119–21
2. Brodsky JB, Macario A, Mark JBD. Tracheal diameter predicts double-lumen tube size: A method for selecting left double-lumen tubes. Anesth Analg 1996;82:861–4
3. Hannallah M, Benumof JL, Silverman PM, Kelly LC, Lea D. Evaluation of an approach to choosing a left double-lumen tube size based on chest computed tomographic scan measurement of left mainstem bronchial diameter. J Cardiothorac Vasc Anesth 1997;11:168–71
4. Chow MYH, Liam BL, Lew TWK, Chelliah RY, Ong BC. Predicting the size of a double-lumen endobronchial tube based on tracheal diameter. Anesth Analg 1998;87:158–60
5. Brodsky JB, Lemmens HJM. Tracheal width and left double-lumen tube size: a formula to estimate left-bronchial width. J Clin Anesth 2005;17:267–70
6. Jeon Y, Ryu HG, Bahk JH, Jung CW, Goo JM. A new technique to determine the size of double-lumen endobronchial tubes by the two perpendicularly measured bronchial diameters. Anaesth Intensive Care 2005;33:59–63
7. Seymour AH, Prakash N. A cadaver study to measure the adult glottis and subglottis: defining a problem associated with the use of double-lumen tubes. J Cardiothorac Vasc Anesth 2002;16: 196–8
8. Campos JH. Current techniques for perioperative lung isolation in adults. Anesthesiology 2002;97:1295–301
9. Slinger P. Lung isolation in thoracic anesthesia, state of the art. Can J Anaesth 2001;48:R1–10
10. Brodsky JB, Lemmens HJM. Left double-lumen tubes: clinical experience with 1,170 patients. J Cardiothorac Vasc Anesth 2003;17:289–98
11. Wilson WC, Benumoff JL. Anesthesia for Thoracic Surgery. In Miller RD (ed). Anesthesia. Philadelphia, Churchill Livingstone, 2005;1877
12. Hurford WE, Alifille PH. A quality improvement study of the placement and complications of double-lumen endobronchial tubes. J Cardiothorac Vasc Anesth 1993;7:517–20
13. Seymour AH, Prasad B, McKenzie RJ. Audit of double-lumen endobronchial intubation. Br J Anaesth 2004;93525–7
14. Fitzmaurice BG, Brodsky JB. Airway rupture from double-lumen tubes. J Cardiothorac Vasc Anesth 1999;13:322–9
15. Tanaka A, Isono S, Ishikawa T, Sato J, Nishino T. Laryngeal resistance before and after minor surgery. Endotracheal tube versus Laryngeal Mask Airways.TM Anesthesiology 2003;99: 247–8
16. Kark AE, Kissin MW, Auerbach R, Meikle M. Voice changes after thyroidectomy: Role of the external laryngeal nerve. BMJ 1984;289:1412–5
17. Domino KB, Posner KL, Caplan RA, Cheney FW. Airway injury during anesthesia: A closed claims analysis. Anesthesiology 1999;91:1703–11
18. Desiderio DP, Burt M, Kolker AC, Fischer ME, Reinsel R, Wilson RS. The effects of endobronchial cuff inflation on double-lumen endobronchial tube movement after lateral decubitus positioning. J Cardiothorac Vasc Anesth 1997;11:595–8
19. Hannallah MS, Gharagozloo F, Gomes MN, Chase GA. 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. Anesth Analg 1998;87: 1027–31
20. Hannallah MS, Benumof JL, Bachenheimer LC, Mundt DJ. The resting volume and compliance characteristics of the bronchial cuff of left polyvinyl chloride double-lumen endobronchial tubes. Anesth Analg 1993;77:1222–6
21. Wagner DL, Gammage GW, Wong ML. Tracheal rupture following the insertion of a disposable double-lumen endotracheal tube. Anesthesiology 1985;63:698–700
22. Kaloud H, Smolle-Juettner FM, Prause G, List WF. Iatrogenic ruptures of the tracheobronchial tree. Chest 1997;112:774–8
23. Lieberman D, Littleford J, Horan T, Unruh H. Placement of left double-lumen endobronchial tubes with or without a stylet. Can J Anaesth 1996;43:238–42
24. Brodsky JB, Benumof JL, Ehrenwerth J, Ozaki GT. Depth of placement of left double-lumen endotracheal tubes. Anesth Analg 1991;73:570–2
25. Yasumoto M, Higa K, Nitahara S, Shono S, Hamada T. Optimal depth of insertion of left-sided double-lumen endobronchial tubes cannot be predicted from body height in below average-sized adult patients. Eur J Anaesthesiol 2006;23:42–4