Adoption and challenges of the 2018 International Society for Heart and Lung Transplantation consensus statement on adult and pediatric airway complications grading system in lung transplantation: A multicenter survey Authors: Maria M. Crespo, Amparo Solé, Christian A. Bermudez, Thomas J. Richards, Zane Mazur, Marie Budev Published: 2026 ABSTRACT Airway complications remain a major cause of morbidity and mortality after cardiothoracic transplantation. The reported incidence of airway ischemic complications continuously varies widely. We conducted an international multicenter survey to better understand the global adoption of the 2018 ISHLT airway grading consensus system and to identify key challenges across centers and areas for potential refinement. Airway complications remain one of the most challenging aspects in lung transplantation, contributing significantly to morbidity, mortality, and reduced quality of life despite major advances in surgical techniques and perioperative management. Their reported incidence varies widely across centers, reflecting not only differences in practice but also the historical absence of standardized definitions and grading systems.1, 2 In response to this variability, the International Society for Heart and Lung Transplantation (ISHLT), Airway Complications Working Group, in 2017, was tasked with the following: (1) to develop a universal definition of the various adult and pediatric airway complications, and (2) to develop a complimentary grading system for these various airway complications. The subsequent 2018 ISHLT Consensus Statement on Adult and Pediatric Airway Complications provided a standardized method for classifying airway complications including malacia, stenosis, ischemic necrosis and dehiscence.3 The aim of the 2018 Consensus document was to provide a consistent definition for each airway complication, with the subsequent goal to improve identification and possibly reduction of complications. This framework represented a major step forward, aiming to standardize reporting, improve communication among clinicians, and enable more consistent research and clinical decision-making. In this context, we conducted a multicenter survey to evaluate the global adoption of the 2018 ISHLT airway grading consensus system, assess its perceived usefulness in clinical practice and research, and identify key challenges and areas for potential refinement to consider in developing future updates to the airway grading consensus system. Materials and methods We conducted an international, cross-sectional survey including multiple-choice, Likert-scale,4, 5, 6 and open-ended items, in reference to the 2018 “ISHLT consensus statement on adult and pediatric airway complications after lung transplant: definitions, grading system, and therapeutics”.3 The survey was sent to a total of 77 lung transplant centers in North American, Europe, Australia/Oceania, and Middle East/Africa. Responses were analyzed mainly from centers currently using the 2018 ISHLT airway grading consensus system. All responses were anonymous and reported in aggregate. The study received Institutional Board approval (UPENN IRB number 825580). Centers were asked to rank clinical parameters important to the use of the 2018 ISHLT airway grading consensus system (Table 1), and to rate their use and utility on a Likert scale (LS) from 1 to 5 (where 1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree). It was also ranked as the challenge to identify airway complications comparing ischemia, necrosis, stenosis, malacia, and dehiscence on a Likert scale from 1 to 5 (where 1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree.Table 12018 ISHLT Adult and Pediatric Airway Complications After Lung Transplant Grading Consensus SystemComplicationCategoryClassificationIschemia and Necrosis (I)Locationa. Perianastomotic (within <1 cm of anastomosis)b. Extending >1 cm to major airways (bronchus intermedius, distal left mainstem)c. Extending >1 cm into lobar or segmental airwaysExtenta. <50% circumferential ischemiab. >50%-100% circumferential ischemiac. <50% circumferential necrosisd. >50%-100% circumferential necrosisDehiscence (D)Locationa. Cartilaginous b. Membranous c. BothExtenta. 0%-25% circumferenceb. >25%-50% circumferencec. >50%-75% circumferenced. >75% circumferenceStenosis (S)Locationa. Anastomoticb. Anastomotic + lobar/segmentalc. Lobar/segmental onlyExtenta. 0%-25% reduction in cross-sectional areab. >25%-50% reductionc. >50% to <100% reductiond. 100% obstructionMalacia (M)Locationa. Perianastomotic (within <1 cm of anastomosis)b. Diffuse (extending beyond anastomosis)Adapted from the 2018 ISHLT consensus statement on adult and pediatric airway complications after lung transplantation. Table 12018 ISHLT Adult and Pediatric Airway Complications After Lung Transplant Grading Consensus SystemComplicationCategoryClassificationIschemia and Necrosis (I)Locationa. Perianastomotic (within <1 cm of anastomosis)b. Extending >1 cm to major airways (bronchus intermedius, distal left mainstem)c. Extending >1 cm into lobar or segmental airwaysExtenta. <50% circumferential ischemiab. >50%-100% circumferential ischemiac. <50% circumferential necrosisd. >50%-100% circumferential necrosisDehiscence (D)Locationa. Cartilaginous b. Membranous c. BothExtenta. 0%-25% circumferenceb. >25%-50% circumferencec. >50%-75% circumferenced. >75% circumferenceStenosis (S)Locationa. Anastomoticb. Anastomotic + lobar/segmentalc. Lobar/segmental onlyExtenta. 0%-25% reduction in cross-sectional areab. >25%-50% reductionc. >50% to <100% reductiond. 100% obstructionMalacia (M)Locationa. Perianastomotic (within <1 cm of anastomosis)b. Diffuse (extending beyond anastomosis)Adapted from the 2018 ISHLT consensus statement on adult and pediatric airway complications after lung transplantation. 2018 ISHLT Adult and Pediatric Airway Complications After Lung Transplant Grading Consensus System Survey questionnaire data were exported from REDCap7 to Excel and imported into the R Environment for Statistical Computing, version 4.5.2. Text strings, including Likert scales,4 were converted to factors or ordered factors, where applicable. Central tendency measures—median, mean—for describing ordered factors were computed using 1-to-5 or 1-to-3 equally-spaced scores. Confidence intervals were calculated using the Wald method. Response proportions for binary outcomes, including the analysis in centers using the ISHLT airway grading consensus system on whether the airway grading system needs revision, were estimated using mixed-effects logistic regression, with a random effect for center to account for clustering of respondents within centers. Similarly, a mixed-effects logistic regression with binomial response and logit link was used to test whether the use of the airway grading system or the need for an airway grading system revision varied significantly across continents or countries, with a random intercept for center and a fixed covariate effect tested using a likelihood-ratio Chi-squared test. Plotting Likert scales was done using plot functions in the R packages Likert and ggplot2. Results Demographics and practice characteristics The survey was sent to a total of 77 lung transplant centers. Responses were received from 58 of 77 (75%) lung transplant centers across 17 countries, including North America (USA and Canada), Australia/Oceania, the Middle East/Africa, and Europe. There were 67 respondents from 58 lung transplant centers, with 6 centers (10.3%) having multiple respondents (1 center with 4 respondents, 1 with 3, and 4 with 2) while 52 centers (89.7%) had 1 respondent. The median category number of respondent years in practice after fellowship completion was >10 years (62.5%; Mean=2.48; SD=0.74). The median category number of lung transplants per year at centers was 51-100 (50%; mean=2.77; SD=0.88). Across centers, bronchoscopy procedures are performed under different organizational models. 29.3% of centers perform bronchoscopies exclusively by lung transplant pulmonologists, while 20.7% are performed by interventional bronchoscopy teams. The majority (50.0%) adopt a hybrid approach, in which pulmonologists, interventional bronchoscopists, and lung transplant surgeons participate in performing bronchoscopy procedures. Of 67 total respondents, 52 reported practices with only adult lung transplants, 5 with only pediatric patients, and 10 with both adult and pediatric patients. Of the 67 respondents, 48 (71.6 %) used the ISHLT airway grading consensus system (38 adult-only programs, 3 pediatric programs, and 3 combined adult and pediatric programs). Usage, utility, and challenges of the ISHLT airway grading consensus system The analysis examines the use of the ISHLT airway grading consensus system at the center level (n=58 centers) rather than the respondent level (n=67 respondents). Overall, 39 of the 58 centers (67.2%) report using the ISHLT airway grading consensus system. The North America (USA and Canada) 30/37 centers (81.1%) had the highest number of respondents, followed by 8/19 European centers (42.1%). Table 2 shows the overall usage patterns of the ISHLT airway grading consensus system across all countries. A total of 7 countries (41.2%) had 100% of their centers using the ISHLT airway grading consensus system, whereas 8 countries (47.1%) had 0% use, and 2 countries (11.8%) had partial use. Centers were classified as 'using the ISHLT airway grading consensus system' if at least 1 respondent from that center reported routine use of the ISHLT airway grading consensus system. Countries in Table 3 are sorted by the percentage of their centers that use the guidelines, with ties broken by the total number of respondents and the country name.Table 22018 ISHLT Airway Grading Consensus System: Overall Usage Patterns Across All CountriesUsage patternNumber of countriesPercentage of countriesTotal centersCountries with 100% center usage741.2%8Countries with 0% center usage*847.1%10Countries with partial center usage**211.8%40Total17100%58*Countries with 0% usage: Belgium, Finland, France, Germany, Italy, Norway, Saudi Arabia, United Kingdom**Countries with partial usage: Spain and USATable 32018 ISHLT Airway Grading Consensus System: Usage by CentersCountryContinentCenters using N-centers usage/N=total centers (%)Total respondentsCanadaNorth America2/2 (100%)2SwedenEurope1/1 (100%)2AustraliaAustralia1/1 (100%)1AustriaEurope1/1 (100%)1DenmarkEurope1/1 (100%)1NetherlandsEurope1/1 (100%)1SwitzerlandEurope1/1 (100%)1USANorth America28/35 (80%)43SpainEurope3/5 (60%)5 Table 22018 ISHLT Airway Grading Consensus System: Overall Usage Patterns Across All CountriesUsage patternNumber of countriesPercentage of countriesTotal centersCountries with 100% center usage741.2%8Countries with 0% center usage*847.1%10Countries with partial center usage**211.8%40Total17100%58*Countries with 0% usage: Belgium, Finland, France, Germany, Italy, Norway, Saudi Arabia, United Kingdom**Countries with partial usage: Spain and USA 2018 ISHLT Airway Grading Consensus System: Overall Usage Patterns Across All Countries *Countries with 0% usage: Belgium, Finland, France, Germany, Italy, Norway, Saudi Arabia, United Kingdom **Countries with partial usage: Spain and USA Table 32018 ISHLT Airway Grading Consensus System: Usage by CentersCountryContinentCenters using N-centers usage/N=total centers (%)Total respondentsCanadaNorth America2/2 (100%)2SwedenEurope1/1 (100%)2AustraliaAustralia1/1 (100%)1AustriaEurope1/1 (100%)1DenmarkEurope1/1 (100%)1NetherlandsEurope1/1 (100%)1SwitzerlandEurope1/1 (100%)1USANorth America28/35 (80%)43SpainEurope3/5 (60%)5 2018 ISHLT Airway Grading Consensus System: Usage by Centers At the center level (n=58), the ISHLT airway grading consensus system was numerically highest among centers where bronchoscopies are performed by the lung transplant pulmonologist (76.5%), followed by centers using a hybrid model (65.5%), and centers where the interventional pulmonary group performs the bronchoscopies (58.3%). However, these differences were not statistically significant (Fisher's exact test, p = 0.54). Results were similar at the respondent level (n = 67), with ISHLT airway grading consensus system usage rates of 73.7%, 71.4%, and 61.5% for the pulmonologist, hybrid model, and interventional groups, respectively (Fisher's exact test, p = 0.78). The probability that a center uses the airway grading system was not significantly associated, at the level of either center (p = 0.54) or respondent (p = 0.78), with who performs transplant bronchoscopies. Usefulness perception of the ISHLT airway grading consensus system: research and management Table 4 and Figure 1 present Likert-scale summary results on the ISHLT airway grading consensus system, useful for research and airway management when airway pathology is identified, based on responses from 48 responders at centers that use the ISHLT airway grading system. Both questions show a positive median response (Agree). Research usefulness had an agreement rate of 68.7% (Mean = 3.81), and management usefulness had an agreement rate of 60.4% (Mean = 3.60). There were very low disagreement rates, 8.3% (research) and 8.3% (management). Values are presented as Count (Percentage). Light blue highlighting indicates positive responses (Agree/Strongly Agree).Table 4Likert-Scale Summary Results for the ISHLT Airway Grading Consensus System Usefulness Perception for Research and Airway ManagementResponse categoryISHLT useful for research N (%)ISHLT useful for management N (%)Strongly Disagree0 (0.0%)1 (2.1%)Disagree4 (8.3%)3 (6.2%)Neutral11 (22.9%)15 (31.2%)Agree23 (47.9%)24 (50.0%)Strongly Agree10 (20.8%)5 (10.4%)StatisticISHLT Useful for ResearchISHLT Useful for ManagementN (sample size)4848Mean3.813.60Standard Deviation0.870.84MedianAgreeAgreeFigure 1Likert-scale responses with central tendency markers useful perception on the ISHLT airway grading consensus system in research and management Table 4Likert-Scale Summary Results for the ISHLT Airway Grading Consensus System Usefulness Perception for Research and Airway ManagementResponse categoryISHLT useful for research N (%)ISHLT useful for management N (%)Strongly Disagree0 (0.0%)1 (2.1%)Disagree4 (8.3%)3 (6.2%)Neutral11 (22.9%)15 (31.2%)Agree23 (47.9%)24 (50.0%)Strongly Agree10 (20.8%)5 (10.4%)StatisticISHLT Useful for ResearchISHLT Useful for ManagementN (sample size)4848Mean3.813.60Standard Deviation0.870.84MedianAgreeAgree Likert-Scale Summary Results for the ISHLT Airway Grading Consensus System Usefulness Perception for Research and Airway Management Likert-scale responses with central tendency markers useful perception on the ISHLT airway grading consensus system in research and management Sixty-eight percent of respondents felt that the ISHLT airway grading system has improved communication between providers and bronchoscopy at their center. Responder’s comments included communication was improved by the following: helping management of airway complications; early identification of complications; helping being more consistent with the diagnosis and severity of airway pathology and assisting with longitudinal follow up; forces people to think about percent involvement and extension; improved consistency in how airways are described; helping to improve communication between pulmonologists and surgeons when interventions are needed. Variability and challenges Most experience low to moderate inter-observer variability in defining and grading airway ischemia and necrosis with the ISHLT airway grading consensus system (87.5%) at their center. Overall, 66.7% found the grading system difficult to use clinically in some areas, and 25% reported no difficulties. Table 5 represents Likert-scale summary response levels, from lowest to highest, for clinical challenges in the diagnosis and grading of airway pathology using the ISHLT airway grading consensus system. Response levels were ordered from lowest to highest difficulty/frequency. Differentiating ischemia from necrosis is the most challenging aspect (47.9% rate as “most challenging”), and grading airway malacia is the second most challenging (33.3% rate as “most challenging”).Table 5Likert-Scale Summary Responses Levels for Clinical Challenges in the Diagnosis and Grading of Airway Pathology Using the 2018 ISHLT Airway Grading Consensus SystemResponse levelDifferentiate airway ischemia from necrosis N=sample size (%)Location and extension of airway ischemia and necrosis N=sample size (%)Least challenging3 (6.2%)8 (16.7%)Somewhat challenging8 (16.7%)14 (29.2%)Challenging5 (10.4%)13 (27.1%)Very challenging9 (18.8%)12 (25.0%)Most challenging23 (47.9%)1 (2.1%) Table 5Likert-Scale Summary Responses Levels for Clinical Challenges in the Diagnosis and Grading of Airway Pathology Using the 2018 ISHLT Airway Grading Consensus SystemResponse levelDifferentiate airway ischemia from necrosis N=sample size (%)Location and extension of airway ischemia and necrosis N=sample size (%)Least challenging3 (6.2%)8 (16.7%)Somewhat challenging8 (16.7%)14 (29.2%)Challenging5 (10.4%)13 (27.1%)Very challenging9 (18.8%)12 (25.0%)Most challenging23 (47.9%)1 (2.1%) Likert-Scale Summary Responses Levels for Clinical Challenges in the Diagnosis and Grading of Airway Pathology Using the 2018 ISHLT Airway Grading Consensus System Perception of the ISHLT airway grading consensus system needs revision The survey also examined responses to the question: “Do you think the current 2018 ISHLT airway grading system needs to be revised?” among 48 respondents from 39 centers that routinely use the grading system. The analysis accounts for clustering of respondents within centers by using a mixed-effects logistic regression model with random center-level intercepts. The model-estimated probability that a respondent believes the system needs revision was 60.4% (crude percentage) and 60.5% (model-estimated percentage), with 95% CI (45.5%-73.8%). Clustering analysis suggests minimal variation between centers. Response rates vary across the 39 centers, ranging from 0% to 100%, supporting revision. The majority of respondents from centers that actively use the ISHLT airway grading consensus system believe it needs to be modified. We also compared North America and Europe to determine whether there are significant regional differences in perceived need for revision of the ISHLT airway grading consensus system. This focused comparison eliminates potential confounding from the single Australian respondent and provides a more robust statistical comparison between the 2 regions with adequate sample sizes. The analysis includes 47 respondents from 38 centers across the 2 main regions (Table 6). There were 38 responders from 30 North American centers and 9 responders from 8 European centers. There were no significant differences in the responses between North America and Europe (p = 0.787). North America (23/38) respondents (60.5%) believe revision is needed [Model: 60.6%] and Europe (5/9) respondents (55.6%) believe revision is needed [Model: 55.6%] with a difference of 4.9 % points (crude)/5.0 % points (model-adjusted); Odds Ratio: 0.814 (95% CI: 0.185-3.593). Europe had slightly lower odds of perceived need for revision, but the difference is not statistically significant. The wide confidence interval for Europe reflects the smaller sample size (n = 9), but the substantial overlap with North America's interval indicates no meaningful difference.Table 6Comparison of North America and Europe Perceptions on The Need for the 2018 ISHLT Airway Grading Consensus System RevisionContinentN=respondersCentersResponse: Yes95% CINorth America383023 (60.5%) [60.6%]43.9%-75.1%Europe985 (55.6%) [55.6%]24.8%-82.6%Totals (N)473828 (59.6%)--Effect (% Difference)----(4.9%) [5.0%]--Effect (Odds ratio, Europe: North America----0.814 (0.185, 3.593)--Likelihood Ratio p-value----0.7868 (NS)--ContinentNCentersResponse: Yes95% CINorth America383023 (60.5%) [60.6%]43.9%-75.1%Europe985 (55.6%) [55.6%]24.8%-82.6%Format: N (crude %) [model-estimated %] Table 6Comparison of North America and Europe Perceptions on The Need for the 2018 ISHLT Airway Grading Consensus System RevisionContinentN=respondersCentersResponse: Yes95% CINorth America383023 (60.5%) [60.6%]43.9%-75.1%Europe985 (55.6%) [55.6%]24.8%-82.6%Totals (N)473828 (59.6%)--Effect (% Difference)----(4.9%) [5.0%]--Effect (Odds ratio, Europe: North America----0.814 (0.185, 3.593)--Likelihood Ratio p-value----0.7868 (NS)--ContinentNCentersResponse: Yes95% CINorth America383023 (60.5%) [60.6%]43.9%-75.1%Europe985 (55.6%) [55.6%]24.8%-82.6%Format: N (crude %) [model-estimated %] Comparison of North America and Europe Perceptions on The Need for the 2018 ISHLT Airway Grading Consensus System Revision Format: N (crude %) [model-estimated %] Of the 67 total respondents, 20 (30%) reported that their program does not use the ISHLT airway grading consensus system; of these, 12 (60%) believe it needs revision. Thus, some centers, mainly in Europe, may refrain from using the airway grading due to perceived shortcomings; we believe it is just as likely that the non-adoption of the airway grading is driven by unfamiliarity. Conclusions We report a summary of the current utilization and perceived challenges of the 2018 ISHLT airway grading consensus system across lung transplant programs in North America, Canada, Europe, Australia, and the Middle East, using a multicenter survey. Overall, 67.2% of centers reported using the ISHLT airway grading system, supporting its broad adoption in clinical practice. Our findings demonstrate a strong agreement that the 2018 ISHLT grading system is (1) useful for research and airway management, (2) improving providers' communication by enabling early identification of airway pathology and supporting clinical decision-making. These findings reinforce the value of a standardized framework for the assessment of airway complications following lung transplantation. Despite the many strengths of the airway grading system, important limitations in its real-world application were identified. 66.7% of respondents found the ISHLT grading difficult in some areas. Differentiating ischemia from necrosis is the most challenging aspect rated as most challenging' and grading airway malacia is the second most challenging rated as most challenging. These challenges likely reflect intrinsic limitations of the current definitions, as well as variability in interpretation across clinicians and centers. Regarding the need for revision or update of the 2018 ISHLT airway grading system, the survey responses were heterogeneous with institutional inconsistency as a few centers (5.2%) showed internal disagreement about revision needs. Overall, 60.4% of respondents believe the current ISHLT airway grading system needs revision in certain areas. The lack of significant difference between North America (USA and Canada) and Europe strengthens the conclusion that the perceived need for ISHLT system revision provides strong evidence for trans-Atlantic consistency in the perceived need for ISHLT grading revision, with as high as 60% agreement in both regions, and suggests universal recognition for revision, supporting global consensus. This survey has several limitations including imbalance in sample size between North America and Europe, and the limited European representation, with only 8 centers. Center clustering was accounted for in the analysis but may still affect generalizability. Our study shows that the current 2018 ISHLT consensus statement on adult and pediatric airway complications after lung transplant: definitions, grading system, and therapeutics has been accepted by the lung transplant community as clinically useful. However, based on the challenges in recognition and differentiating certain pathologies, updates and revisions of the current ISHLT airway grading consensus system will help to address these challenges and have better improvement in early pathology recognition and management. Disclosure statement M.C. receives royalties from UpToDate. M.B. receives consulting fees from Lung Bioengineering Advisory Board, Transplant Growth, Transmedics Advisory Board, Kamada Speaker Bureau, and Sanofi Advisory Board. C.B., A.S., T.R., and M.Z. have no disclosures. This project was partially supported by the Stern Foundation, University of Pennsylvania. CRediT Authorship Contribution Statement MC, MB: conceptualization, methodology, protocol design, survey preparation, data analysis, writing original draft. AS and CAB: manuscript review and editing and editing. TJR: formal analysis and statistical analysis, manuscript review. ZM: building survey in REDCAP, data curation. Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Marie Budev receives consulting fees from Lung Bioengineering Advisory Board, Transplant Growth, Transmedics Advisory Board, Kamada Speaker Bureau, and Sanofi Advisory Board. The remaining authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.