Share this article on:

Multicenter Experience With Electromagnetic Navigation Bronchoscopy for the Diagnosis of Pulmonary Nodules

Jensen, Kurt W. MD*; Hsia, David W. MD*; Seijo, Luis M. MD; Feller-Kopman, David J. MD; Lamb, Carla MD§; Berkowitz, David MD; Curran-Everett, Douglas PhD*,¶; Musani, Ali I. MD*

Journal of Bronchology & Interventional Pulmonology: July 2012 - Volume 19 - Issue 3 - p 195–199
doi: 10.1097/LBR.0b013e3182616ece
Original Investigations

Background: Physicians are increasingly encountering lung nodules in their practice, and tissue diagnosis is often required. Conventional bronchoscopic sampling yields a range from 14% to 69% depending on the nodule size and location within the lung. We aimed to evaluate the diagnostic yield of electromagnetic navigation bronchoscopy (ENB) in multiple centers and to determine what factors affect the yield of ENB.

Methods: A retrospective analysis of 92 consecutive ENB procedures at 5 centers was carried out. Data were collected on patient demographics, nodule characteristics, complications, type of samples obtained, diagnosis, and follow-up studies. Variables were analyzed to determine as to which factors had an impact on the diagnostic yield with multiple logistic regression analysis.

Results: Ninety-two patients underwent EMB at 5 centers between December 2008 and October 2009. The average nodule size was 2.61 cm (SD 1.42) at a distance of 1.81 cm (SD 1.32) from the pleural surface. The overall yield for ENB-guided sampling of pulmonary nodules was 65% (60/92). The ENB yield for nodules ≤2 versus >2 cm in size was significantly less after controlling for the distance from the pleura (50% vs. 76%, respectively; P=0.01). The distance from the pleura did not affect the ENB diagnostic yield after controlling for nodule size (P=0.92). The lobar location of the nodule also did not affect the diagnostic yield (P=0.59).

Conclusions: The diagnostic yield of ENB-guided sampling of pulmonary nodules is impacted by the nodule size, but not by the distance from the pleura or the lobar location.

*National Jewish Health, Denver, CO

IIS-Fundacion Jimenez Diaz-CIBERES, Madrid, Spain

Department of Medicine, Division of Pulmonary Medicine, Johns Hopkins University, Baltimore, MD

§Lahey Clinic, Boston, MA

Emory University Hospital, Atlanta, GA

Colorado School of Public Health, University of Colorado Denver, Denver CO

Disclosure: L.M.S. has received 10,000 euros as a clinical course director for electromagnetic navigation courses and has been a nonpaid member of superDimension’s European advisory board. C.L. was an educational consultant for superDimension in 2007. A.I.M. has received less than $10,000 as a consultant and a clinical course instructor for electromagnetic navigation bronchoscopy course organized by superDimension Inc. For the remaining authors, there is no conflict of interest or other disclosures.

Reprints: Kurt W. Jensen, MD, Pulmonary Cunsultants 316 MLK Jr Way, Suite 401 Tacoma, WA 98405 (e-mail: kjensen@pulmonaryconsultants.org).

Received April 24, 2012

Accepted May 25, 2012

Pulmonary nodules are a frequent finding on computed tomography (CT) scans of the chest. It is estimated that 1 in 500 chest x-rays will demonstrate a pulmonary nodule.1 Even higher incidences are reported on CT scans,2,3 and the majority of smokers will have nodules seen on CT imaging.4 More than 150,000 patients with lung nodules are assessed by physicians in the United States annually and this number will likely increase with the advent of lung cancer screening using CT as suggested by a recent study (not published) showing a 20% decline in mortality in patients older than 55 years of age, who have a history of 30 pack years of smoking.5

For lesions <8 mm in size, the likelihood of malignancy is <1%.6 Current recommendations are to follow these lesions with serial CT scans at varying time intervals depending on the patient’s risk factors for lung cancer.7 However, the risk of malignancy is 18% for lesions 8 to 20 mm in diameter and 50% for lesions >20 mm.6 Thus, many patients with lung nodules will require a diagnostic procedure to evaluate for malignancy.

The routine bronchoscopic diagnostic yield for malignant disease varies with the size and location of the lesion. The sensitivity of bronchoscopy ranges from 34% (range, 5% to 76%) for lesions <20 mm to 63% (range, 31% to 82%) for lesions >20 mm in size.8 New navigation technologies have been developed recently to improve the diagnostic yield of bronchoscopy in the evaluation of pulmonary nodules. These include electromagnetic navigation bronchoscopy (ENB), virtual bronchoscopy with and without navigational aide, and radial probe ultrasound. ENB creates a virtual bronchoscopic image of the patient’s airways derived from a CT scan of their chest. The patient is then placed in an electromagnetic field generated by a board placed under the patient. The virtual bronchoscopy image is then aligned with the patient’s actual anatomy, allowing a steerable sensor probe to be navigated to the lesion under virtual real-time guidance. After reaching the lesion, a guide catheter (extended working channel) is left at the lesion and allows samples to be obtained. Other navigational systems provide a virtual bronchoscopic image with guidance provided by a predetermined route, not with real-time positional navigation as with ENB.

In this study, we aim to evaluate the diagnostic yield of ENB in peripheral pulmonary nodules in routine clinical practice. We also aim to determine what factors affect the diagnostic yield of ENB and the complication rate of ENB.

Back to Top | Article Outline

MATERIALS AND METHODS

Data were collected retrospectively by chart reviews of the most recent 20 consecutive ENB cases performed from December 2008 to October 2009 at each of the 5 participating institutions. All procedures were performed with the superDimension ENB system (superDimension, Inc.; Minneapolis, MN) as per each institution’s standard protocol either under moderate procedural sedation or using deep sedation. Informed consent was obtained before each procedure. The Institutional Review Board at National Jewish Health approved the protocol (HS-2509).

Data were collected on patient demographics, nodule size, lobar location, distance from pleura, methods of sampling utilized (ie, needle aspiration, brushing, lavage, and/or transbronchial biopsy), mean FEV1, complications, type of sedation, and diagnoses obtained.

Nodule characteristics were determined using the CT scan used for ENB planning. The nodule size was recorded as the largest diameter on axial imaging. The distance from the pleura was measured from the nodule to the nearest parietal pleural surface. Only the lobar location, and not the segment of the lobe, was recorded. Pathology results from ENB or subsequent procedures were used to determine the final diagnoses. If no specific diagnosis was obtained by bronchoscopy, surgical biopsy results or stablility for 6 months on radiographic follow-up were used to determine benignity.

The diagnostic yield was calculated as the number of successful diagnoses by ENB divided by the total number of procedures multiplied by 100. ENB was classified as successful if it resulted in a specific malignant or benign diagnosis, or if a nonspecific benign biopsy was subsequently proven to be benign by further sampling, or if subsequent CT imaging demonstrated nodule stability ≥6 months after ENB biopsy. If a nonspecific benign result was obtained on ENB biopsy but no follow-up sampling or CT scan was performed, the procedure was deemed unsuccessful.

The SAS/STAT software package, version 9.2, of the SAS System for Windows XP (copyright © 2002 to 2008, SAS Institute Inc.) was used for all statistical analyses. The primary method of analysis to determine as to which factors had an impact on the diagnostic yield was multiple logistic regression analysis. In the initial model, we included 5 variables for theoretical reasons: the nodule size, the distance from the pleura, the lobar location, the type of anesthesia, and the clinical site. Because the lobar location, the type of anesthesia, and the clinical site were not important predictors in estimating the diagnostic yield, the final multiple logistic regression model included the nodule size and the distance from the pleura.

Back to Top | Article Outline

RESULTS

Data from a total of 92 patients were analyzed. The mean age was 67 years (SD 13) and 52% were women (Table 1). The mean FEV1 was 2.05 L (SD 0.97). The average nodule size was 2.61 cm (SD 1.42) and was 1.81 cm (SD 1.32) from the pleural surface. Eight patients’ data were excluded from analysis because of insufficient follow-up to determine the benign nature of nonspecific biopsy results.

The diagnostic yield for ENB-guided sampling of pulmonary nodules was 65% (60/92). Two ENB-guided biopsies demonstrated nonspecific inflammation, but were deemed successful as the lesions resolved on subsequent CT imaging. There was no significant difference between the institutions in the diagnostic yield (P=0.70; Tables 2 and 3). A nonsignificant increase in the diagnostic yield was observed with moderate procedural sedation versus propofol (logistic regression: P=0.62) although ENB cases using propofol involved significantly smaller nodules [1.95 (SD 0.84) vs. 2.69 (SD 1.39) cm; t test (unequal variance assumption): P=0.005]. Moreover, the ENB yield for nodules ≤2 cm in size versus those >2 cm in size was significantly worse after controlling for the distance from the pleura (50% vs. 76%, respectively; logistic regression: P=0.01): these estimates were obtained from the parameter estimates for nodule size and distance from the pleura that resulted from the final logistic regression model. The distance of the nodule from the pleura did not have any effect on the diagnostic yield of ENB-guided sampling after controlling for the nodule size (66% vs. 65%, respectively; logistic regression: P=0.92): these estimates were obtained from the parameter estimates for nodule size and distance from the pleura that resulted from the final logistic regression model. The lobar location of the nodule also did not affect the diagnostic yield (P=0.59; Table 3). The type of sampling method and the number of sampling methods used did not affect the diagnostic yield (data not shown).

The overall complication rate was 4%, with 3 pneumothoraces and 1 episode of bleeding, but none required hospitalization. The pneumothorax rate was not significantly different for nodules ≤2 cm from the pleural surface versus nodules >2 cm from the pleural surface (1/52 vs. 2/40, respectively; Fisher exact test: P=0.58).

Back to Top | Article Outline

DISCUSSION

The routine diagnostic bronchoscopic yield for malignant disease is 34% for nodules <2 cm and 63% for nodules >2 cm in size. ENB increases the diagnostic yield to a reported range of 59% to 77% (Table 4).9–15 Our study represents a real-world assessment of the accuracy of ENB in that the procedures were performed by ENB-trained or training pulmonologists and not in a study setting. The nodule size also affected the diagnostic yield in our study, with a yield of 76% for nodules >2 cm in size and 50% for nodules ≤2 cm in size. In fact, the only significant determinant of diagnostic yield in our study was the size of the nodule sampled. The nodule’s lobar location, the distance from the nearest pleural surface, and the mode of sedation utilized for the procedure did not affect the diagnostic yield. The effect of nodule size on the biopsy yield is in contrast to the study by Seijo et al.16 In their study, the nodule size did affect the diagnostic yield, but the effect did not reach statistical significance.

Recently, Seijo et al16 demonstrated that the presence of a bronchus sign, an airway terminating in or going through a lesion, significantly affects the yield of ENB. We did not evaluate the effect of this factor as we did not have data available on the presence or the absence of a bronchus sign for all of our cases. However, utilizing the results of our study and the study of Seijo and colleagues, we can now likely begin to further improve on the diagnostic yield of ENB for pulmonary nodules by choosing nodules with favorable characteristics.

Another new technology available for improving bronchoscopic biopsy of peripheral lung lesions is radial ultrasound, with reported yields of 46% to 77%.17–20 Radial endobronchial ultrasound offers the benefit of real-time confirmation that the guide catheter has been appropriately placed in or near the target lesion. In contrast, ENB relies on virtual guidance to the target lesion identified on CT; this will have an inherent degree of error depending on the accuracy of the registration process aligning the patient’s anatomy to the virtual CT anatomy. However, ENB offers the potential advantage of a steerable guide, which can navigate the guide catheter through the distal airway branching that might otherwise be impossible to maneuver into. Ideal navigation bronchoscopy would theoretically utilize the advantages of both systems, and this method has been shown to be superior (yield of 88%) to either method used alone.21 Radial probe ultrasound was utilized in a few procedures involved in our analysis, but the subgroup was too small to allow for meaningful analysis.

Finally, the overall complication rate in our study was low (4%). The complication included 3 pneumothoraces and 1 episode of bleeding. None of these required hospitalization.

Back to Top | Article Outline

CONCLUSIONS

ENB offers an improvement over standard bronchoscopy techniques in the diagnosis of pulmonary nodules, and is primarily dependent on the nodule size and not the location. With accumulating evidence on the factors that affect the yield of ENB, such as the nodule size and the presence of the bronchus sign, pulmonologists should be able to further improve the yield of ENB with appropriate case selection and reduce the number of procedures patients may need to undergo in an attempt to diagnose their pulmonary nodule(s).

Back to Top | Article Outline

REFERENCES

1. Tan B, Flaherty K, Kazerooni E, et al. The solitary pulmonary nodule. Chest. 2003;123:89S–96S
2. Henschke C, Yankelevitz D, Libby D, et al. International early lung cancer action program investigators. N Engl J Med. 2006;355:1763–1771
3. Menezes R, Robert H, Paul N, et al. Lung cancer screening using low-dose computed tomography in at-risk individuals: the Toronto experience. Lung Cancer. 2010;67:177–183
4. Swensen SJ, Silverstein MD, Ilstrup DM, et al. The probability of malignancy in solitary pulmonary nodules: application to small radiographically indeterminate nodules. Arch Intern Med. 1997;157:849–855
5. . Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011;365:395–409
6. Midthun DE, Swensen SJ, Jett JR, et al. Evaluation of nodules detected by screening for lung cancer with low dose spiral computed tomography. Lung Cancer. 2003;41(suppl 2):S40
7. MacMahon H, Austin JHM, Gamsu G, et al. Guidelines for management of small pulmonary nodules deceted on CT scans: a statement from the Fleischner Society. Radiology. 2005;237:395–400
8. Rivera MP, Mehta AC. Initial diagnosis of lung cancer: ACCP evidence-based clinical practice guidelines (2nd edition). Chest. 2007;132:131S–148S
9. Gildea TR, Massone PJ, Karnak D, et al. Electromagnetic navigation diagnostic bronchoscopy. Am J Respir Crit Care Med. 2006;174:982–989
10. Eberhardt R, Anantham A, Ernst A, et al. Multimodality bronchoscopic diagnosis of peripheral lung lesions. Am J Respir Crit Care Med. 2007;176:36–41
11. Makris D, Scherpereel A, Leroy S, et al. Electromagnetic navigation diagnostic bronchoscopy for small peripheral lung lesions. Eur Respir J. 2007;29:1187–1192
12. Eberhardt R, Anantham D, Herth F, et al. Electromagnetic navigation diagnostic bronchoscopy in peripheral lung lesions. Chest. 2007;131:1800–1805
13. Lamprecht B, Porsch P, Pirich C, et al. Electromagnetic navigation bronchoscopy in combination with PET-CT and rapid on-site cytopathologic examination for diagnosis of peripheral lung lesions. Lung. 2009;187:55–59
14. Wilson DS, Bartlett RJ. Improved diagnostic yield of bronchoscopy in a community practice: combination of electromagnetic navigation system and rapid on-site evaluation. J Bronchol. 2007;14:227–232
15. Schwarz Y, Greif J, Becker HD, et al. Real-time electromagnetic navigation bronchoscopy to peripheral lung lesions using overlaid CT images. Chest. 2006;129:988–994
16. Seijo LM, de Torres JP, Lozano MD, et al. Diagnostic yield of electromagnetic navigation bronchoscopy is highly dependent on the presence of a bronchus sign on CT imaging: results from a prospective study. Chest. 2010;138:1316–1321
17. Kurimoto N, Miyazawa T, Okimasa S, et al. Endobronchial ultrasonography using a guide sheath increases the ability to diagnose peripheral pulmonary lesions endoscopically. Chest. 2004;126:959–965
18. Yoshikawa M, Sukoh N, Yamazaki K, et al. Diagnostic value of endobronchial ultrasonography with a guide sheath for peripheral pulmonary lesions without x-ray fluroscopy. Chest. 2007;131:1788–1793
19. Kikuchi E, Yamazaki K, Sukho N, et al. Endobronchial ultrasonography with guide-sheath for peripheral pulmonary lesions. Eur Respir J. 2004;24:533–537
20. Eberhard R, Ernst A, Herth FJF. Ultrasound-guided transbronchial biopsy of solitary pulmonary nodules less than 20 mm. Eur Respir J. 2009;34:1284–1287
21. Eberhardt R, Anantham D, Ernst A, et al. Multimodality bronchoscopic diagnosis of peripheral lung lesions: a randomized controlled trial. Am J Respir Crit Care Med. 2007;176:36–41
Keywords:

bronchoscopy; lung nodule; lung cancer; biopsy; diagnostic bronchoscopy; lung cancer diagnosis; electromagnetic navigation bronchoscopy

© 2012 Lippincott Williams & Wilkins, Inc.