Clinical Application Of Gemstone Spectral CT-imaging On Identification/Diagnosis Of Different Pathological Types Of Cancers And Their Metastatic Lymph Nodes

Objective To investigate the clinical value of the gemstone spectral imaging (GSI) indifferential diagnosis/identification of various types of lung, gastric and esophagealcancers with different tissue origin and pathological features, as well as their respectivemetastatic lymph nodes through comparative analysis of the characteristic CT spectralimaging parameters from different tissue-derived and/or different pathological types oftumors and metastatic lymph nodes.Methods From March2012to October2012,30patients with lung cancer, gastriccancer or esophageal cancer underwent GSI examination. Among them,7cases of smallcell lung cancer (10lymph nodes),9cases of lung adenocarcinoma (11lymph nodes),7cases of lung squamous cell carcinoma (7lymph nodes), and7cases of gastric oresophageal cancer (13lymph nodes) were included. In the polychromatic andmonochromatic imaging from these cancers and metastatic lymph nodes, the CT valuesat different keV (40keV-140keV, with10keV interval), iodine contents and Effective-Zwere measured, respectively. Spectral attenuation curves and their curve slopes werethen obtained. The slope of spectral curve is calculated through dividing the CT valuedifference between two keV points by energy (keV) difference between the two keVpoints. Herein, we selected40keV and80keV points, and the Slope40keV-80keV=(HU40keV-HU80keV)/40. The region of interest (ROI) of the metastatic lymph node inthe CT imaging was selected with the standard of short diameter greater than10mm,and necrotic areas avoided. All data were compared between groups and independentsamples t test was used for statistical analysis. Results There were no significant differences for the CT values at different keV, iodinecontents, Effective-Z and the spectral curve slopes among tumor primary lesions of thelung cancers (including small cell lung cancers, lung adenocarcinomas, and lungsquamous cell carcinomas). There were also no significant differences for all of the CTspectral imaging parameters between tumor primary lesions from small cell lungcancers or lung adenocarcinomas, and their metastatic lymph nodes. However, the CTvalues at40keV-110keV, iodine contents and the spectral curve slopes of tumor primarylesions from small cell lung cancers or lung adenocarcinomas had statisticallysignificant differences when compared with those of the metastatic lymph nodes fromgastric or esophageal cancers, respectively. There were no significant differences for allthe CT spectral imaging parameters between the metastatic lymph nodes from small celllung cancers and those from lung adenocarcinomas; whereas, the CT values at40keV-130keV and all the other spectral imaging parameters of the metastatic lymphnodes from small cell lung cancers or lung adenocarcinomas had statistically significantdifferences in comparison with those of the metastatic lymph nodes from gastric oresophageal cancers, respectively (iodine content P<0.01, others P<0.05); the CT valuesat different keV and iodine contents of the metastatic lymph nodes from small cell lungcancers or lung adenocarcinomas also had statistically significant differences whencompared with those of the metastatic lymph nodes from lung squamous cellcarcinomas respectively [CT value P<0.05, in low-energy (keV) zone P<0.01; iodinecontent: lung squamous cell carcinoma metastatic lymph nodes v.s. small cell lungcancer metastatic lymph nodes t=3.024, P=0.004, lung squamous cell carcinomametastatic lymph nodes v.s. lung adenocarcinoma metastatic lymph nodes t=2.942,P=0.005]. There were significant differences for the CT values at40keV-110keV andiodine contents of the metastatic lymph nodes from lung squamous cell carcinoma whencompared with those of the tumor primary lesions of all types of lung cancers or their metastatic lymph nodes. The CT values at80keV-100keV and iodine contents of themetastatic lymph nodes from lung squamous cell carcinomas were also statisticallysignificantly different when compared with those of the metastatic lymph nodes fromgastric or esophageal cancers.Conclusion Comparative analysis of the characteristic CT spectral imaging parameters(e.g., CT value, iodine content, Effective-Z and spectral curve slope) from tumorprimary lesions and their metastatic lymph nodes has certain values in clinical guidancefor differential diagnosis/identification of various types of tumors with differenttissue/organ origin and pathological characteristics, as well as their correspondingmetastatic lymph nodes.