DIAGNOSTIC YIELD OF NEXT GENERATION SEQUENCING IN THE EVALUATION OF PLEURAL EFFUSIONS

Abstract

TOPIC: Disorders of the Pleura TYPE: Original Investigations PURPOSE: Next Generation Sequencing (NGS) is a massive parallel sequencing technology that is becoming increasingly important in clinical microbiology. Its use has the potential to change workflows in the evaluation of multiple infectious syndromes, particularly those that have been historically difficult to diagnose based on conventional culture. This retrospective study aimed to compare the diagnostic yield of NGS and bacterial culture in the evaluation of pleural effusions. METHODS: Subjects were recruited from the Pleural practice at Mayo Clinic Rochester from January 2017 to August 2019 with IRB approval. Pleural fluid aliquots were collected at the time of thoracentesis or thoracostomy tube placement. Clinical characteristics, including patient demographics, pleural fluid chemistries, Gram stain and culture results, and antimicrobial usage patterns prior to pleural fluid sampling, were obtained from the electronic medical record. NGS was performed by microbiology laboratory personnel blinded to the clinical data.Subjects were stratified into groups based on the likelihood that the sampled pleural effusion was secondary to empyema. We defined proven cases as those having a consistent clinical presentation with empyema in addition to either positive Gram stain or bacterial culture and/or the presence of frank pus at the time of pleural fluid aspiration. Probable cases were defined as exudates by Light’s criteria, with a pH less than 7.2 in association with clinical evidence of infection (fever, leukocytosis, and/or elevated CRP). Unlikely cases did not meet the proven or probable criteria. Unclassifiable cases were missing information needed to determine whether they qualified as proven or probable cases. Proven and probable cases were considered to be true empyemas when calculating sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of NGS.Agreement between NGS and bacterial culture data was categorized as total agreement (negative testing or same organism detected by both methods), partial agreement (single organism detected by one method and multiple organisms detected by other method), or no agreement. RESULTS: Our study included 47 patients with a mean age of 70 and majority male population. The breakdown of cases included 9 proven empyemas, 3 probable empyemas, 17 unlikely empyemas, and 18 cases that were unclassifiable. The diagnostic yield of NGS in diagnosing empyemas was as follows: sensitivity 92%, specificity 74%, PPV 55%, and NPV 96%. The diagnostic yield of bacterial culture was not able to be characterized to the same degree, as bacterial culture was not ordered on all effusions included in our study. However, we were able to determine that the sensitivity of bacterial culture in diagnosing empyema was 58%. Of the 34 cases that had available culture and NGS data, there was total agreement between the two methods in 67.7% of cases, partial agreement in 8.8% of cases, and discordance in 23.5% of cases. CONCLUSIONS: The diagnostic evaluation of pleural effusions that are potentially infectious in nature can be augmented by NGS given its high NPV. CLINICAL IMPLICATIONS: Further study is needed to determine the most appropriate patient population in whom NGS will enhance the diagnosis and management of pleural effusions without overly increasing the detection of clinically insignificant bacterial burden in the pleural space. DISCLOSURES: No relevant relationships by Dagny Anderson, source=Web Response No relevant relationships by Eva Carmona, source=Web Response