Many studies [9, 14–16] have shown that TDs are an important prognostic factor for GC. According to Anup et al. , the prognoses of patients with TDs in the pT1–3 category were identical to those in the pT4a category, indicating that TDs may be better handled as a type of serosal invasion. As a result, en bloc resection of the primary tumor is critical, and adjuvant treatment should always be considered if TDs are found. In addition, Sun et al.  observed that the number of TDs was positively correlated with worsening prognosis. As the number of TDs increased, the five-year survival rate of patients decreased gradually. Many other studies [18, 19] have emphasized that TDs differ from MLNs because the former have a worse effect on prognosis. Therefore, accurate diagnosis of perigastric TDs and MLNs in gastric cancer before treatment is conducive to the selection of appropriate treatment regimens and the assessment of prognosis. Currently, there are four main types of imaging examinations for gastric cancer patients: MDCT, endoscopic ultrasonography (EUS), magnetic resonance imaging (MRI), and positron emission tomography (PET)/CT . MDCT is widely used in gastric cancer patients due to its fast scanning speed, high spatial resolution, and stable image quality , but there are relatively few studies on the use of MDCT to evaluate perigastric TDs in gastric cancer. We hope that the preoperative CT evaluation of perigastric TDs in gastric cancer can help doctors determine treatment strategies more accurately and select the best type of surgery for their patients.
In our study, we used contrast-enhanced CT to distinguish between perigastric TDs and MLNs in gastric cancer. The contrast-enhanced CT images of 36 patients with gastric cancer were analyzed, and the morphological characteristics and parameters of contrast-enhanced CT in the pathologically confirmed TDs (30) and MLNs (55) were evaluated. The results showed that morphology and size could be used to differentiate TDs from MLNs. Most TDs have irregular morphology, and some have lobulation and spiculation. This irregular morphology may be caused by the proliferation, migration, and accumulation of tumor cells and their heterogeneous growth in different directions, or it may be related to the formation of TDs. Puppa et al.  believed that TDs may be a new type of destructive metastatic invasion involving blood vessels and perivascular structures. Most MLNs have a smooth appearance and regular morphology (mostly round or oval). In terms of pathological manifestation, in MLNs, the tumor cells are encapsulated by the lymph nodes due to the immune response and are not able to grow beyond the lymph node capsule. Regular morphology is an important appearance factor that can be used to distinguish TDs from MLNs. However, it has been reported that MLNs can also have irregular morphology, perhaps because the tumor grows in the lymph nodes and gradually invades them .
In this study, the long-axis and short-axis diameters of TDs were larger than those of MLNs. It is possible that TDs contain more tumor cells than MLNs and grow faster. Atre et al.  showed that on MRI, TDs adjacent to rectal cancer have an irregular morphology and larger long- and short-axis diameters than MLNs, a finding that is consistent with the results of this study. In our study, the aspect ratios of TDs and MLNs were not significantly different, perhaps because TDs tend to be round or oval overall despite having an irregular morphology. In addition, the cCAVs in each phase were significantly higher in the TD group than in the MLN group, possibly due to the high density of tumor cells and the abundance of tumor blood vessels in TDs, characteristics that can also be used to differentiate between TDs and MLNs.
To date, research on perigastric TDs in gastric cancer has mainly focused on their prognostic value and how to reasonably include them in TNM staging [25, 26]. The present study is the first to predict TDs in gastric cancer by using CT imaging. With the continuous innovation and development of CT technology, the diagnostic information obtained using CT is no longer limited to morphological manifestations and CT values of lesions but also incorporates other functions. such as dual energy CT (DECT), CT-perfusion (CTp) and radiomics, etc. [27–31]. The application of these methods to MLNs in gastric cancer has been studied previously, but the prediction of gastric cancer TDs by these methods remains to be verified. In addition, in an imaging study of TDs in rectal cancer, Lord et al.  reported that TDs are related to nodules in the intramesorectal venous pathways on MRI images that mainly connect to the venous branches and do not connect to the primary tumor. This new prognostic marker is included in the MRI report and is called mrTDs. Lord et al.  also showed that mrTDs and extramural venous invasion (mrEVMI) on MRI images have higher prognostic accuracy than the current clinical TNM staging of rectal cancer. However, because gastric veins are more abundant, slender, and tortuous than rectal veins and because contrast-enhanced CT has a poorer ability than MRI to display gastric veins, the conclusions of Lord et al. could not be verified in the present work. In future studies, we will combine dual DECT, CTp, MRI and radiomics to make more accurate judgments regarding perigastric TDs in gastric cancer, predict the disease condition of patients, and inform accurate and reasonable treatments for patients in clinical practice.
This study has the following limitations. (1) Because all of the nodules examined in this study had a diameter greater than 5 mm, nodules with diameters less than 5 mm may have been missed. Even when nodules have a diameter less than 2 mm, the possibility of TDs and MLNs cannot be excluded [33, 34]. (2) Some of the available data were lost because it was difficult to correspond some of the TDs or MLNs with CT images. The difficulty of one-to-one correspondence in surgery increases with decreasing nodule volume, resulting in a certain bias in the selection of nodules. In addition, the sample size was relatively small. The selected measurement parameters failed to show all of the imaging characteristics of TDs. In future studies, we will need to expand the sample size and enrich the measurement parameters to obtain more abundant and more accurate data. (3) No CT examinations performed using DECT, CTp, or radiomics are available for comparison with enhanced CT imaging in the evaluation or detection of TDs.
In conclusion, this study demonstrates that contrast-enhanced MDCT scanning is beneficial for predicting TDs in patients with gastric cancer prior to surgery. The lesion morphology and size observed by MDCT and the cCAVs in the noncontrast phase and each contrast-enhanced phase have reference value for differentiating between perigastric TDs and MLNs. TD-positive gastric cancer patients have poor prognosis. Hence, predicting and identifying TDs through preoperative contrast-enhanced CT can guide clinicians in selecting more suitable treatment regimens to maximize the benefits of treatment.