The combined use of serum Raman spectroscopy and D dimer testing for the early diagnosis of acute aortic dissection

Abstract

ObjectivesAcute aortic dissection (AAD) is an extremely life-threatening medical emergency, often misdiagnosed in its early stages, resulting in prolonged wait times for rescue. This study aims to identify potential serum biomarkers that can assist in the accurate diagnosis of AAD and effectively differentiate it from other conditions causing severe chest pain. MethodsA total of 122 patients with AAD and 129 patients with other severe chest pain disorders were included in the study. Serum samples were analyzed by measuring the peak intensities of Raman spectra. For each measurement, the Raman spectrum was accumulated by two accumulations (3 s per acquisition). Logistic regression and nomogram models were developed using these peak intensities as well as D-dimer levels to predict the occurrence of AAD. The clinical utilities of these models were assessed through receiver operating characteristics (ROC) curve analysis, net reclassification improvement (NRI), integrated discrimination improvement (IDI), and decision curve analysis (DCA) in both training and internal test cohorts. ResultsThe D-dimer levels of AAD patients were significantly increased, as well as higher intensities at specific Raman peaks, including 505 cm−1, 842 cm−1, 947 cm−1, 1254 cm−1, 1448 cm−1, and 1655 cm−1 when compared to non-AAD patients. Conversely, decreased intensities were observed at Raman peaks such as 750 cm−1, 1004 cm−1, 1153 cm−1, 1208 cm−1, and 1514 cm−1 in AAD patients. Least absolute shrinkage and selection operator regression analysis on the training cohort identified eight potential predictors: D-dimer along with intensity measurements at peaks such as 505 cm−1, 750 cm−1, 1153 cm−1, 1208 cm−1, 1254 cm−1, 1448 cm−1, and 1655 cm−1. The combination of these eight potential predictors demonstrated a good discriminatory performance, with an area under the curve (AUC) value of 0.928 in the training cohort and an AUC of 0.936 in the internal test cohort, outperforming the use of D-dimer alone. Furthermore, DCA curve analysis revealed that leveraging this combination of eight potential predictors would provide substantial net benefits for clinical application. Moreover, this combination significantly augmented discrimination power, as evidenced by a continuous NRI of 39.8 % and IDI of 9.95 % in the training cohort, as well as a continuous NRI of 27.1 % and IDI of 9.95 % in the internal test cohort. ConclusionsThe employment of this combination of eight potential predictors effectively rules out AAD to a greater extent. This study presents a promising diagnostic strategy for early detection using optical diagnostic techniques such as Raman spectroscopy.