Complications like acute cellular rejection (ACR) and infection are known risk factors for the development of chronic lung allograft dysfunction, impacting long-term patient and graft survival after lung transplantation (LTx). Differentiating between complications remains challenging and time-sensitive, highlighting the need for accurate and rapid diagnostic modalities. We assessed the ability of exhaled breath analysis using an electronic nose (eNose) to distinguish between ACR, infection, and mechanical complications in LTx recipients (LTR) presenting with suspected complications. LTR with suspected complications and subsequently proven diagnosis underwent exhaled breath analysis using an eNose. Supervised machine learning was used to assess the eNose’s ability to discriminate between complications. Next, we determined the added value of the eNose measurement on top of standard clinical parameters. In 90 LTR, 161 measurements were performed during suspected complications, with 84 proven diagnoses. The eNose could distinguish between ACR, infection, and mechanical complications with 74% accuracy, and ACR and infection with 82% accuracy. Combining eNose measurements with standard clinical parameters improved diagnostic accuracy to 88% (P =.0139), with 94% sensitivity and 80% specificity. Exhaled breath analysis using eNose technology is a promising, noninvasive, diagnostic modality for distinguishing LTx complications, enabling timely diagnosis and interventions.

中文翻译：

---

电子鼻区分肺移植受者并发症的能力


急性细胞排斥反应 （ACR） 和感染等并发症是慢性肺同种异体移植物功能障碍发展的已知危险因素，影响肺移植 （LTx） 后患者和移植物的长期存活率。区分并发症仍然具有挑战性且具有时间敏感性，这凸显了准确和快速诊断方式的必要性。我们评估了使用电子鼻 （eNose） 进行呼气分析的能力，以区分出现疑似并发症的 LTx 受者 （LTR） 的 ACR、感染和机械并发症。使用 eNose 对疑似并发症和随后确诊的 LTR 进行呼气分析。监督机器学习用于评估 eNose 区分并发症的能力。接下来，我们确定了 eNose 测量在标准临床参数之上的附加值。在 90 LTR 中，在疑似并发症期间进行了 161 次测量，其中 84 例确诊。eNose 可以区分 ACR、感染和机械并发症，准确率为 74%，ACR 和感染的准确率为 82%。将 eNose 测量与标准临床参数相结合，诊断准确性提高到 88% （P =.0139），灵敏度为 94%，特异性为 80%。使用 eNose 技术进行呼气分析是一种很有前途的无创诊断方式，可用于区分 LTx 并发症，实现及时诊断和干预。