Spectral CT Analysis of Solitary Pulmonary Nodules for Differentiating Malignancy from Benignancy: The Value of Iodine Concentration Spatial Distribution Difference

Abstract

Objective The objective is to assess the value of spatial distribution difference in iodine concentration between malignant and benign solitary pulmonary nodules (SPNs) by analyzing multiple parameters of spectral CT. Methods Sixty patients with 39 malignant nodules and 21 benign nodules underwent chest contrast CT scans using spectral imaging mode during pulmonary arterial phase (PP), arterial phase (AP), and venous phase (VP). Iodine concentrations of proximal and distal regions in pulmonary nodules on iodine-based material decomposition images were recorded. Normalized iodine concentration (NIC) and the differences in NIC between the proximal and the distal regions (dNIC) were calculated. The two-sample t-test and Mann–Whitney U-test were performed to compare the multiple parameters generated from spectral CT between malignant and benign nodules. Receiver operating characteristic (ROC) curves were generated to calculate sensitivity and specificity. Results NIC in the proximal region (NICpro) and NIC in the distal region (NICdis) between malignant and benign nodules at AP (NICpro, P=0.012; NICdis, P=0.024), and VP (NICpro, P=0.005; NICdis, P =0.004) were significantly different. NICpro at PP (P = 0.037) was also found significantly different between malignant and benign nodules; however, no significant differences were found in NICdis at PP (P = 0.093). In addition, the dNIC of malignant nodules was significantly higher than that of benign ones at PP (median and interquartiles (0.31, 0.11, 0.57 versus -0.26, -0.5, -0.1); p≤0.001), AP (mean dNIC, 0.093 ±0.094 versus -0.075±0.060; p≤0.001), and VP (mean dNIC, 0.171±0.137 versus -0.183±0.127; p≤0.001). The sensitivity and specificity (93%, 95%, respectively) of dNIC during VP were higher than other parameters, with a threshold value of -0.07. Conclusions Spectral CT imaging with multiple parameters such as NICpro, NICdis, and dNIC may be a new method for differentiating malignant SPNs from benign ones.

The application of dual-layer spectral detector computed tomography in solitary pulmonary nodule identification

“…According to our results, the value of NIC was greater in the malignant group than in the benign group, and NIC VP was superior in differentiating malignant from benign nodules. This result was similar to the findings of previous studies that demonstrated the sensitivity and specificity of NIC VP to be in the ranges of 93-93.8% and 85.7-95%, respectively (4,(17)(18)(19). Xiao et al (18) focused on the value of kVp-switching dual-energy CT in SPN diagnosis.…”

“…Furthermore, Zhang et al (4) reported that NIC was a more valuable index than λ HU and that CT numbers, especially in the VP, demonstrated a sensitivity and specificity of 93.8% and 85.7%, respectively. Another study regarding the distribution difference of iodine concentration between malignant and benign SPNs found that the differences in NIC (dNIC) between the proximal and the distal regions in the VP were optimal in the characterization of SPNs, with a sensitivity and specificity of 93% and 95%, respectively (17).…”

“…Poorly differentiated carcinoma is characterized by more aggressive growth, with a lower micro-vessel density, leading to a lower value of NIC (35). Previous investigations showed that the use of iodinerelated quantitative parameters from source-based dualenergy CT (fast kVp switching) improved the specificity for differentiating benign from malignant pulmonary nodules (17). However, the radiation dose for source-based dual-energy CT is higher than that for conventional CT, and an a priori decision is also required to select the scanning mode (4,19).…”

Non-malignant lung pathologies showing 18F-Fluorodeoxyglucose positron emission tomography/computed tomography positivity

“…The definitive diagnosis is made using a biopsy or surgical procedure if the biopsy fails. In recent years, data show the delayed 18 F-FDG PET/CT method, with the aim of reducing misinterpretation and comparison of early and late SUVmax values, or the new calculation method, the relative activity distribution (rAD) method, can increase the differentiation between malignant and benign in PET/CT imaging (24,25). There are also studies that use different radionuclide agents such as 11 C-1-methionine and 18 F-fluoro-alphamethyl-tyrosine (26,27).…”

Visual assessment of 18F-FDG metabolic spatial distribution improves the differential diagnosis of indeterminate pulmonary nodules and masses with high FDG uptake

“…The proximal region of lesion was defined as the area near the ipsilateral hilar, and the area far from the ipsilateral hilar was defined as the distal region. [12] A 5-point scoring system was used to interpret the characteristics of FMSD, which represent the likelihood of the lesion being benign or malignant (Table 1). When the characteristics of FMSD in lesions are not obvious, the metabolic threshold can be manually adjusted on transverse PET images to assist the interpretation.…”