View Abstract

A1099 - A Global Evaluation of Agreement on the Decision to Perform Surgical Lung Biopsy Between Pulmonologists in the Setting of Diffuse Lung Disease
Author Block: S. L. F. Walsh1, T. M. Maher2, M. R. Kolb3, V. Poletti4, R. Nusser5, L. Richeldi6, C. Vancheri7, M. L. Wilsher8, A. Antoniou9, J. Behr10, E. Bendstrup11, K. K. Brown12, L. Calandriello13, T. J. Corte14, V. Cottin15, B. Crestani16, K. R. Flaherty17, I. Glaspole18, J. C. Grutters19, Y. Inoue20, M. Kokosi2, Y. Kondoh21, V. Kouranos22, M. Kreuter23, K. A. M. Johannson24, E. Judge25, B. J. Ley26, G. Margaritopoulos27, F. J. Martinez28, M. Molina Molina29, A. Morais30, H. Nunes31, G. Raghu32, C. J. Ryerson33, M. Selman34, P. Spagnolo35, H. Taniguchi36, D. Taniguchi37, H. Taniguchi36, S. Tomassetti38, D. Valeyre39, M. Wijsenbeek40, W. Wuyts41, D. Hansell42, A. U. Wells43; 1Radiology, King College London, London, United Kingdom, 2Interstitial Lung Disease Unit, Royal Brompton Hospital, London, United Kingdom, 3Mc Master Univ, Hamilton, ON, Canada, 4Department of Thoracic Diseases, Morgagni-Pierantoni Hospital, Forli, Italy, 5Summit Hospital, Oakland, CA, United States, 6Respiratory Medicine, Catholic University of the Sacred Heart, Rome, Italy, 7Inst of Respiratory Diseases, Catania, Italy, 8Auckland City Hosp, Auckland 1, New Zealand, 9Univerity of Crete, Heraklio, Greece, 10Internal Medicine V, University of Munich, Comprehensive Pneumology Ctr., Munchen, Germany, 11Department of Respiratory Diseases and Allergology, Aarhus University Hospital, Aarhus, Denmark, 12Natl Jewish Hlth, Denver, CO, United States, 13Institute of Radiology, Università Cattolica del Sacro Cuore,, Rome, Italy, 14Respiratory Medicine, Royal Prince Alfred Hospital, Sydney, Australia, 15Louis Pradel Hosp, Univ of Lyon, Lyon (Bron), France, 16Service de Pneumologie, Hopital Bichat, Paris, France, 17Univ of Michigan, Ann Arbor, MI, United States, 18Alfred Hospital, Melbourne, Australia, 19Dept of Pulmonology, St Antonius Hosp, Nieuwegein 3435 CM, Netherlands, 20Natl Hosp Org Kinki-Chuo Chest Med Ctr, Osaka, Japan, 21Tosei Gen Hosp, Seto Aichi 489-8642, Japan, 22Interstitial Lung Disease, Royal Brompton Hospital, London, United Kingdom, 23Center for interstitial and rare lung diseases, Thoraxklinik, University of Heidelberg, Heidelberg, Germany, 24Medicine, University of Calgary, Calgary, AB, Canada, 25Royal Brompton Hospital, London, London, United Kingdom, 26Medicine, UCSF, San Francisco, CA, United States, 27Royal Brampton Hospital, London, United Kingdom, 28Weill Cornell Medical College, New York, NY, United States, 29University Hospital of Bellvitge-IDIBELL-CIBERE
Objectives
To evaluate the interobserver agreement on surgical lung biopsy decisions in the setting of diffuse lung disease (DLD) among a large group of pulmonologists from many different countries.
Methods
404 pulmonologists (including 34 interstitial lung disease [ILD] experts) from 57 different countries independently evaluated a cohort of 60 cases of ILD. For each case, a differential diagnosis, likelihoods for each diagnosis and the decision to perform or not to perform surgical lung biopsy was recorded. Interobserver agreement on surgical lung biopsy decisions in pulmonologist subgroups was calculated using the weighted Kappa coefficient.
Results
24240 pulmonologist-case reviews were completed. Overall interobserver agreement on the decision to perform surgical lung biopsy between expert pulmonologists was fair (n=34, 0.29, IQR 0.20-0.39), poor between the remaining pulmonologists (n=370, 0.14, IQR 0.04-0.25). Stratified based on hospital setting, interobserver agreement on the decision to perform surgical lung biopsy was poor between university pulmonologists (n=288, 0.16, IQR 0.05-0.26), and poor between community pulmonologists (n=116, 0.13, IQR 0.03-0.24). Interobserver agreement improved between the experts on subgroup analysis of patients younger than 55 and greater than 75 years and who had a DLco of less than 50 %predicted or greater than 70 %predicted (0.39, IQR=0.38-0.52) but only marginally improved between the remaining pulmonologists (0.15, IQR 0.02-0.31). Factors that independently predicted the decision to proceed to surgical lung biopsy were, first-choice diagnostic confidence, IPF diagnostic likelihood category based on Ryerson et al. AJRCCM 2017 Apr 17. doi: 10.1164/rccm.201702-0400PP. [Epub ahead of print], patient age and patient %predicted DLco.
Conclusions
Interobserver agreement on surgical lung biopsy decisions is poor to fair among pulmonologists of different levels of expertise and from different hospital settings. Disagreement is highest in patients between the ages of 55-75 and when the %predicted DLco lies between 50-70%. Guideline recommendations on biopsy decisions in diffuse lung disease are needed.