Work

Survey of Lung Transplant Community’s Views on Primary Graft Dysfunction

Year: 2016

Type: article

Abstract: To the Editor: We recently published a predictive rule for primary graft dysfunction (PGD) after lung transplantation (1Shah RJ Diamond JM Cantu E Objective estimates improve risk stratification for primary graft dysfunction after lung transplantation.Am J Transplant. 2015; 15 (et al): 2188-2196Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar. While there is a range of published PGD incidence, there are no published data on “acceptable” PGD incidence. We sought to gauge the opinions of the transplant community on the acceptable incidence and predicted individual risk of PGD. We generated a questionnaire, posted to the International Society of Heart and Lung Transplantation (ISHLT) Pulmonary Council and emailed to participants in the Lung Transplant Outcomes Group (LTOG). Responses were collected anonymously using SurveyMonkey (Palo Alto, CA) (Figure 1) (2SurveyMonkey, Inc. [accessed 2015 Apr 12]. Available from www.surveymonkey.com.Google Scholar. Of the 68 respondents, 49 (72%) were pulmonologists and 14 (21%) were surgeons. The median response to, “What is an acceptable center-specific percent incidence of severe, grade 3, primary graft dysfunction?” was 12% (range 0, 45%) with similar responses from pulmonologists (12%) and surgeons (10%). The median response to, “At what predicted individual risk of PGD in a specific recipient would you not offer a patient lung transplantation?” was 50% (range 15%, 90%), with no difference between pulmonologists and surgeons. Of the 68 respondents, 51 (75%) felt a predictive model would influence the decision to proceed with transplant. The median response to, “At what predicted individual risk of PGD in a specific recipient would you change patient management (ex. restricted donor quality, early use of ECMO, etc.)?” was 40% (range 5%, 80%), with a lower threshold for surgeons compared to pulmonologists (25% vs. 40%). Of the 68 respondents, 64 (94%) felt a predictive model for PGD would assist in determining whether the transplant surgical plan should be altered. Respondents were internally consistent; those believing predictive models would not be useful for altering transplant plans also responded that modeling would not be useful for refusing transplantation. Risk-averse responders were consistent with lower acceptable PGD incidences and lower thresholds for refusing or changing transplant plans. The survey has limitations. While the number of responses was small and the distribution of LTOG versus ISHLT responses was unknown secondary to the anonymous nature of the survey, the ISHLT discussion group adds international representation. We did not record respondent transplant center, and there may be clustering of opinions regarding PGD by centers, potentially adding bias. The survey is also unable to capture lack of uniformity in PGD grading, the uncertainty in negative and positive predictive values inherent in predictive algorithms, or the impact of prediction on program-level decision making. In summary, we present the first survey of transplant practitioners’ opinions regarding PGD and the utility of PGD prediction. Responses indicated that while variability exists, there was near uniform belief that PGD is unavoidable. The reported incidence of PGD fell within the incidence of severe PGD in clinical research (3Christie JD Bellamy S Ware LB Construct validity of the definition of primary graft dysfunction after lung transplantation.J Heart Lung Transplant. 2010; 29 (et al): 1231-1239Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar,4Diamond JM Lee JC Kawut SM Clinical risk factors for primary graft dysfunction after lung transplantation.Am J Resp Crit Care Med. 2013; 187 (et al): 527-534Crossref PubMed Scopus (417) Google Scholar. Responses indicate that a predictive algorithm would be a useful tool for preparing patients for transplant surgery. Patients at high risk for PGD likely have high lung allocation scores, and avoiding these patients based on predictive risk could increase waitlist mortality. PGD prediction may be utilized for identifying low- and high-risk patients and for designing clinical trials for PGD therapies. Conception and design: JD, RS, EC, MP, JC. Acquisition of data: JD, RS. Analysis and interpretation of data: JD, RS, EC, MP, JC. Drafting/revising the manuscript for important intellectual content: JD, RS, EC, MP, JC. Final approval of the version to be published: JD, RS, EC, MP, JC. This study was supported by National Institutes of Health grants R01 HL087115, R01 HL081619, K24 HL115354, K23 HL121406, K23 HL116656, R01 HL114626, R01 HL096845, Robert Wood Johnson grant AMFDP 70640, and by a grant from Actelion Pharmaceuticals US, Inc. The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

Source: American Journal of Transplantation

DOI: 10.1111/ajt.13552

Authors: J.M. Diamond, R.J. Shah, E. Cantu, M.K. Porteous, J.D. Christie

Institutions: University of Pennsylvania, University of California, San Francisco

Language: English

Topic: Transplantation: Methods and Outcomes

Subfield: Surgery

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: hybrid

Estimated APC paid: $4,300