Volatile organic compound profiling to explore primary graft dysfunction after lung transplantation

Abstract

Primary graft dysfunction (PGD) is a major determinant of morbidity and mortality following lung transplantation. Delineating basic mechanisms and molecular signatures of PGD remain a fundamental challenge. This pilot study examines if the pulmonary volatile organic compound (VOC) spectrum relate to PGD and postoperative outcomes. The VOC profiles of 58 bronchoalveolar lavage fluid (BALF) and blind bronchial aspirate samples from 35 transplant patients were extracted using solid-phase-microextraction and analyzed with comprehensive two-dimensional gas chromatography coupled to time-of-flight mass spectrometry. The support vector machine algorithm was used to identify VOCs that could differentiate patients with severe from lower grade PGD. Using 20 statistically significant VOCs from the sample headspace collected immediately after transplantation (< 6 h), severe PGD was differentiable from low PGD with an AUROC of 0.90 and an accuracy of 0.83 on test set samples. The model was somewhat effective for later time points with an AUROC of 0.80. Three major chemical classes in the model were dominated by alkylated hydrocarbons, linear hydrocarbons, and aldehydes in severe PGD samples. These VOCs may have important clinical and mechanistic implications, therefore large-scale study and potential translation to breath analysis is recommended.

New Aspects of Lung Transplantation: A Narrative Overview Covering Important Aspects of Perioperative Management

“…Similarly, complement activation fragments are detected in the broncho-alveolar lavage within 24 h after lung transplantation and are increased in patients with PGD [ 65 ]. Even more interesting, because earlier, is the analysis of pulmonary volatile organic compounds from bronchial washing and aspirates obtained immediately after surgery as recently reported by Stefanuto et al in 35 lung transplant recipients, of whom ten will develop a grade 3 PGD [ 66 ]. Of note, Sage et al proposed an inflammation score based on IL-6 and IL-8 levels in perfusate of ex vivo lung perfusion to identify lungs more likely to develop primary graft dysfunction at 72-h post-transplant [ 67 ], Keller et al showed that levels of percentage donor-derived cell-free DNA on Day 1 are higher in PGD patients than non-PGD patients ( p = 0.01) [ 68 ], and Chacon-Alberty identified a unique protein pattern in patients who did or did not develop grade 3 PGD at T48–72 h [ 69 ].…”

Recent Advances in GC×GC and Chemometrics to Address Emerging Challenges in Nontargeted Analysis

“…Various supervised ML methods have also been implemented to peak tables produced from GC×GC separations. For example, Stefanuto et al utilized a support vector machine (SVM) model to classify the severity of primary graft dysfunction (PGD) in lung transplant patients (Figure B) . To define, SVM is a supervised machine learning method that aims to maximize the separation between two classes while minimizing misclassifications .…”

“…dysfunction (PGD) in lung transplant patients (Figure 8B). 108 To define, SVM is a supervised machine learning method that aims to maximize the separation between two classes while minimizing misclassifications. 109 As shown in Figure 8Bi, this study analyzed a peak table containing 386 entries generated from HS-SPME-GC×GC-TOFMS separations of lung fluid collected from patients with grade 3 PGD (PGD3) and lower grades of PGD (PGD0−2).…”

Comprehensive two-dimensional gas chromatography with mass spectrometry: an advanced bioanalytical technique for clinical metabolomics studies

“…Advances in GC×GC for metabolomics demonstrate the potential for very high-resolution analysis of human metabolomes, which can support the detection of complex biomarkers of diseases and yield insights into the biochemical processes and/or disease mechanisms. Particularly, Hill's [90][91][92][93][94] and Focant's [95][96][97][98][99][100] groups have contributed significantly to such applications, highlighting salient features of GC× GC for deciphering human metabolomes. In this review, selected recent applications illustrative of the use of and rationale for GC×GC with electron ionization mass spectrometry (EIMS) (Table 1) in the last 10 years (2012-2021) will be discussed to exemplify the potential of clinical metabolomics for early diagnosis, treatment intervention and elucidating the pathogenesis of human diseases.…”