Liu W, Kong DM, An JK, Song LT. Analysis of the clinical value of gemstone spectral computed tomography imaging in the preoperative assessment of colorectal cancer. World J Gastrointest Surg 2025; 17(8): 105391 [DOI: 10.4240/wjgs.v17.i8.105391]
Corresponding Author of This Article
Li-Tao Song, Department of Radiology, Zibo Central Hospital, No. 54 Gongqingtuan West Road, Zibo 255000, Shandong Province, China. peterhot123@163.com
Research Domain of This Article
Radiology, Nuclear Medicine & Medical Imaging
Article-Type of This Article
Retrospective Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Author contributions: Liu W prepared the manuscript; Liu W and Song LT designed the study; Liu W, Kong DM, and An JK collected the data; Liu W and Song LT analyzed the data; and all authors read and approved the final manuscript.
Institutional review board statement: This study was approved by the Medical Ethics Committee of Zibo Central Hospital.
Informed consent statement: As the study used anonymous and pre-existing data, the requirement for the informed consent from patients was waived.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
Data sharing statement: No additional data are available.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Li-Tao Song, Department of Radiology, Zibo Central Hospital, No. 54 Gongqingtuan West Road, Zibo 255000, Shandong Province, China. peterhot123@163.com
Received: April 8, 2025 Revised: April 29, 2025 Accepted: June 13, 2025 Published online: August 27, 2025 Processing time: 139 Days and 3.9 Hours
Abstract
BACKGROUND
The diagnostic accuracy for detecting metastatic lymph nodes in colorectal cancer (CRC) remains suboptimal. To address this limitation, our study investigates the potential of gemstone spectral computed tomography imaging (GSI) to improve diagnostic accuracy in lymph node metastasis (LNM) assessment.
AIM
To extensively investigate the clinical utility of GSI in the preoperative assessment of CRC.
METHODS
The subject population included 200 patients with CRC who were admitted to Zibo Central Hospital from January 2022 to December 2023. All patients underwent dual-phase contrast-enhanced scans in the arterial and venous phases using GSI before surgical intervention. During the research, meticulous quantification was conducted regarding the number of patients with CRC with LNM as well as the exact count of metastatic lymph nodes. Moreover, for both metastatic and non-metastatic lymph nodes, the short diameter at the maximum cross-sectional area (covering the axial, sagittal, and coronal planes), morphological features (including manifestations such as margin blurring, aggregation, and enhancement), and spectral parameters in the arterial and venous phases [specifically iodine concentration (IC), normalized IC (NIC), and the slope of the spectral curve (λHU)] were measured and recorded, and a comparative analysis was conducted. The diagnostic efficacy of each index with differences was systematically assessed using the receiver operating characteristic (ROC) curve. Concurrently, receiver operating characteristic curves were constructed for LNM screening based on the short diameter at the maximum cross-sectional area of lymph nodes and each spectral parameter in the arterial and venous phases.
RESULTS
The area under the curve of GSI for diagnosing LNM in patients with CRC can reach 0.897, with sensitivity, specificity, and accuracy of 92.59%, 85.87%, and 89.50%, respectively. A total of 265 lymph nodes were analyzed from the 200 participants with CRC, with metastatic lymph nodes accounting for 56.60%. Compared with non-metastatic lymph nodes, the short diameters of metastatic lymph nodes in the axial, sagittal, and coronal planes were significantly increased, whereas the IC values in the arterial and venous phases, the NIC value in the arterial phase, and the λHU values in the arterial and venous phases were significantly decreased. The short axial, sagittal, and coronal diameters, arterial-phase IC, venous-phase IC, arterial-phase NIC, arterial-phase λHU, and venous-phase λHU for diagnosing metastatic lymph nodes demonstrated area under the curve values of 0.631, 0.681, 0.659, 0.862, 0.808, 0.831, 0.801, and 0.706, respectively.
CONCLUSION
GSI exhibits substantial clinical significance in the preoperative assessment of CRC. Among the parameters assessed, the arterial-phase IC demonstrates the most outstanding diagnostic performance, effectively improving the diagnostic efficacy for preoperative LNM in CRC.
Core Tip: To improve the diagnostic accuracy of metastatic lymph node detection in colorectal cancer (CRC), this study emphasizes the potential of gemstone spectral computed tomography imaging as a promising solution. Systematic assessment confirms that gemstone spectral computed tomography imaging demonstrates high diagnostic efficacy in detecting lymph node metastasis in patients with CRC. Among the spectral parameters analyzed across arterial and venous phases, iodine concentration in the arterial phase demonstrates the greatest diagnostic performance. These results provide novel clinical information, supporting optimized surgical decision-making and improved monitoring strategies for patients with CRC with lymph node metastasis.
Citation: Liu W, Kong DM, An JK, Song LT. Analysis of the clinical value of gemstone spectral computed tomography imaging in the preoperative assessment of colorectal cancer. World J Gastrointest Surg 2025; 17(8): 105391
Colorectal cancer (CRC) is a tumor that originates from the abnormal proliferation of colonic glandular epithelial cells, which predominantly occurs in the colon or rectum. Its etiopathogenesis is intricately associated with several factors, including dietary composition, lifestyle patterns, familial genetic predisposition, and chronic inflammatory states[1,2]. Global epidemiological data on CRC indicates that in 2020 alone, CRC constituted 10.0% of the 19.3 million newly diagnosed cancer cases globally and 9.4% of the 10 million cancer-related deaths[3]. Notably, the median age of patients with CRC is 58 years. The risk of developing CRC incrementally increases with advancing age, reaching its apex among individuals aged ≥ 80 years. However, recognizing that a non-negligible number of CRC cases occur in the adolescent population is imperative[4,5]. Currently, the therapeutic approach for patients with CRC primarily revolves around an individualized multimodal treatment paradigm, with surgery at its core complemented by an array of adjunctive modalities such as chemotherapy, radiotherapy, molecular-targeted therapy, and immunotherapy[6]. The implementation of precise preoperative assessment for patients with CRC is of paramount importance. It enables surgeons and other relevant attending physicians to establish surgical strategies that are both scientifically sound and highly efficient[7]. In particular, the preoperative assessment of metastatic lymph nodes is not merely intimately associated with the patient’s prognosis but also serves as a pivotal determinant in deciding whether adjuvant chemotherapy is required postoperatively[8]. The diagnostic accuracy of metastatic lymph nodes remains suboptimal in the current clinical diagnosis of CRC[9]. This study aims to conduct an in-depth investigation of this issue, striving to improve the accuracy of preoperative assessment of metastatic lymph nodes in patients with CRC and provide useful references for clinical practice.
Gemstone spectral computed tomography (CT) imaging (GSI) represents an advanced imaging modality that leverages the GE Spectral CT AW4.7 postprocessing workstation for meticulous image postprocessing and precise measurement. Employing multi-planar reconstruction and the delineation of the region of interest (ROI), GSI enables the extraction of comprehensive tumor-related information[10]. Its mono-energy imaging provides distinct advantages, generating superior image quality and more data than conventional CT images. Specifically, it can generate images depicting the density and distribution of base materials, along with 40-140 keV multi-parameter single-energy images. Moreover, based on the acquired spectral curves, GSI enables the calculation of the effective atomic number of lesions or tissues[11]. Chuang et al[12] revealed that the application of GSI in colon cancer has contributed to improving the diagnostic accuracy regarding the degree of tumor differentiation. Similarly, Hong et al[13] demonstrated that this imaging technique has exhibited its versatility in the diagnosis, classification, and prognostic surveillance of gastrointestinal malignancies, thereby facilitating accurate preoperative staging, efficacious discrimination between benign and malignant tumors, and effective assessment of the therapeutic response and prognosis of lesions. Notably, a preliminary empirical investigation has revealed that GSI, through the integration of normalized iodine concentration (NIC) and axial short diameter, significantly improved the diagnostic precision for differentiating metastatic and non-metastatic lymph nodes in rectal cancer[14].
This study primarily aimed to determine the clinical significance of GSI in the preoperative assessment of CRC, evaluating its diagnostic value for patients with lymph node metastasis (LNM), and analyzing the differences between metastatic and non-metastatic lymph nodes by obtaining the short diameters of the largest sections, morphological characteristics, and spectral parameters in arterial and venous phases. Furthermore, a quantitative analysis is conducted to determine the diagnostic value of the short diameters of the maximum cross-sectional areas and the venous-phase spectral parameters in identifying LNM.
MATERIALS AND METHODS
General information
The research cohort in this study involved 200 patients with CRC admitted to Zibo Central Hospital from January 2022 to December 2023. The sample size for this study was identified through power calculations based on the area under the curve (AUC). This study enrolled 200 participants, which substantially exceeds the minimum required sample size of 144 cases. This generates 90% statistical power (1-β = 0.9) at a significance level of α = 0.05 to detect a clinically meaningful AUC difference of ≥ 0.15, thereby ensuring adequate power to demonstrate the superior diagnostic performance of GSI.
Inclusion criteria: All patients were confirmed with CRC based on fiberoptic endoscopy or postoperative pathological assessment[15]. All participants underwent GSI before surgical intervention. Patients received no preoperative anti-tumor therapies or surgical procedures. None of them had contraindications to magnetic resonance imaging examination. Patients were aged ≤ 80 years and reported no perianal surgery history. Patients had complete clinical data.
Exclusion criteria: This study excluded patients with a frail constitution, rendering them unable to endure hypotonic enemas; those exhibiting hypersensitivity to iodine-based contrast agents; patients presenting with contraindications to 654-2, including but not limited to glaucoma, pyloric obstruction, intestinal obstruction, and prostatic hypertrophy; cases where the spectral image quality failed to meet the requisite diagnostic standards; patients whose lesions contained two pathological types.
Examination protocol
Three days pre-assessment, patients were advised to follow a low-fat, -fiber, and -protein, or residue-free diet. Starting 8-12 hours pre-assessment, they were required to fast and abstain from drinking. Additionally, they were prescribed polyethylene glycol for bowel cleansing. Fasting was maintained on the day of assessment. Before the scan, patients were instructed to drink adequate water and hold their urine to achieve a moderately distended bladder. Once they felt the urge to urinate, they were instructed to inform the CT staff immediately. Subsequently, 10 mg of 654-2 was intramuscularly administered. This was performed to minimize intestinal peristalsis and prevent motion artifact formation. Using an automatic enema machine, 1000-1500 mL of warm normal saline was infused into the colorectum within 3-4 minutes, with the exact volume adjusted based on the patient’s comfort and tolerance, ensuring appropriate colorectal dilation. The scan was performed using the GSI energy-spectrum scanning mode. With automatic tube current adjustment, this mode enabled the instantaneous switching between high- and low-energy tube voltages (140 kV and 80 kV) within 0.5 milliseconds. The slice thickness was set at 5 mm, the pitch at 0.984, and the rotation time at 0.6 seconds per revolution. The abdominal aorta was monitored, and the triggering threshold was set at 150 HU. Arterial- and venous-phase scans, both performed in the GSI mode, were initiated at 5.4 seconds and 20 seconds after triggering, respectively. The scanning area extended from the diaphragmatic dome to the pelvic floor. Ioversol, with a 320 mgI/mL concentration, was used as the contrast agent. The total volume of the contrast agent was calculated at a rate of 1.2 mL/kg of body weight. A CT-compatible high-pressure syringe, with a consistent injection rate of 3.0 mL/second, was used for administration.
Surgical and pathological procedures
All surgical pathological specimens from patients were promptly immersed in 10%-13% neutral formalin within 1 hours of collection for a duration of 12-48 hours. Sampling was meticulously executed for CRC lesions along the long axis of the intestinal wall, perpendicular to its surface. Samples were taken separately based on the size, texture, depth of invasion, color, etc., of the tumor, and a full-thickness tissue slice was specifically retrieved from the deepest invasion site. These samples were then fixed in paraffin and sectioned into slices with a 4 μ thickness. Lymph nodes were sampled identically, ensuring that the sampling process included each lymph node. Two experienced pathologists, each holding an attending physician or higher position, independently and blindly reviewed all the prepared slices under both high- and low-magnification microscopes to identify the presence or absence of LNM. In the event of any discrepancies in their assessments, the pathologists engaged in in-depth discussions to reach a consensus.
Image analysis and processing
All image postprocessing and measurement were conducted on the GE Spectral CT AW4.7 postprocessing workstation (GSI-viewer). Based on the postoperative pathology, lymph nodes were categorized into metastatic and non-metastatic groups. The short diameters of the lymph nodes were measured along the axial, sagittal, and coronal planes utilizing multi-planar reconstruction. Meanwhile, characteristics, such as the lymph nodes’ margins, aggregation patterns, and enhancement uniformity, were meticulously documented. At a mono-energy level of 60 kV, the cross-section that most clearly depicted the largest or relatively large portion of the lymph node was designated as the ROI. The iodine concentration (IC) of the targeted lymph node, as well as that of the abdominal aorta or iliac artery at the same section, was measured. Subsequently, the NIC was calculated using the formula, NIC = IC of the lesion/IC of the abdominal aorta or iliac artery at the same section. Additionally, the slope of the spectral curve (λHU) was computed using the formula, λHU = (CT value at 40 kV - CT value at 100 kV)/60. The diameter of each ROI was maintained at < 5 mm, with careful avoidance of necrotic areas. Substantial efforts were exerted to ensure the consistency of the size, shape, and position of the selected ROI between the arterial and venous phases. The ROI at the largest cross-section was measured three times for each lesion, and the mean value was finally adopted for analysis. Radiologists independently assessed the GSI parameters while remaining blinded to pathological results. Conversely, pathologists assessed LNM without access to any prior imaging interpretations. To maintain objectivity, a third-party statistician, uninvolved in either the radiological or pathological assessments, performed the final comparative analysis using fully de-identified datasets.
Statistical analysis
In this research, measurement data that follow a normal distribution are expressed as mean ± SD. An independent t-test is applied to compare two samples. Categorical data are presented as n (%), and a test is used to assess the categorical data between the two groups. To assess the diagnostic efficiency, the receiver operating characteristic (ROC) curve is used to calculate the AUC. A P-value of < 0.05 indicated statistical significance.
RESULTS
Analysis of general characteristics of patients with CRC with and without LNM
No notable differences were found in terms of gender, age, smoking history, alcoholism history, and family medical history between patients with CRC without and with LNM (P > 0.05; Table 1).
Table 1 General data analysis of colorectal cancer patients with and without lymph node metastasis, n (%).
General data
Without lymph node metastasis (n = 108)
With lymph node metastasis (n = 92)
χ2/t
P value
Sex
0.136
0.712
Male
58 (53.70)
47 (51.09)
Female
50 (46.30)
45 (48.91)
Age (year), mean ± SD
60.31 ± 11.26
61.79 ± 8.55
1.032
0.303
Smoking history
2.170
0.141
Without
72 (66.67)
52 (56.52)
With
36 (33.33)
40 (43.48)
Alcoholism history
3.414
0.065
Without
79 (73.15)
56 (60.87)
With
29 (26.85)
36 (39.13)
Family medical history
0.545
0.461
Without
86 (79.63)
77 (83.70)
With
22 (20.37)
15 (16.30)
Diagnostic significance of GSI for LNM in CRC
Among the 200 patients with CRC, pathological diagnosis confirmed that 108 cases (54.00%) exhibited no LNM, whereas 92 (46.00%) cases demonstrated lymph node involvement. Employing GSI, 113 (56.50%) cases were identified as non-metastatic, and 87 (43.50%) were diagnosed with LNM. After calculation, the sensitivity, specificity, and accuracy of GSI in detecting LNM in patients with CRC were 92.59%, 85.87%, and 89.50%, respectively. Notably, AUC reached 0.897. Detailed information is presented in Table 2 and Figure 1.
Figure 1 Receiver operating characteristic curve of gemstone spectral computed tomography imaging for the diagnosis of lymph node metastasis in colorectal cancer.
GSI: Gemstone spectral computed tomography imaging; AUC: Area under the curve.
Table 2 Gemstone spectral computed tomography imaging diagnosis of lymph node metastasis in colorectal cancer.
GSI
Pathological staging
Total
Without Lymph node metastasis
With lymph node metastasis
Without lymph node metastasis
100
13
113
With lymph node metastasis
8
79
87
Total
108
92
200
Short diameter of the maximum cross-sectional area of metastatic and non-metastatic lymph nodes
The analysis in this study included 265 lymph nodes from 200 patients with CRC. Among them, 150 were metastatic, and 115 were non-metastatic lymph nodes. The short diameters of metastatic lymph nodes in the axial, sagittal, and coronal views were markedly greater than those of non-metastatic lymph nodes (P < 0.001; Table 3).
Table 3 Short diameter of the maximum cross-sectional area of metastatic and non-metastatic lymph nodes, mean ± SD.
Short diameter of the largest section (cm)
Metastatic lymph nodes (n = 150)
Non-metastatic lymph nodes (n = 115)
t
P value
Axial short diameter
1.25 ± 0.70
0.94 ± 0.33
4.389
< 0.001
Sagittal short diameter
1.17 ± 0.58
0.80 ± 0.20
6.547
< 0.001
Coronal short diameter
1.17 ± 0.62
0.86 ± 0.28
4.985
< 0.001
Morphological characteristics of metastatic and non-metastatic lymph nodes
Metastatic lymph nodes were more prone to present with blurred margins, aggregation, and uneven enhancement than non-metastatic lymph nodes (P < 0.05; Table 4).
Table 4 Morphological characteristics of metastatic and non-metastatic lymph nodes, n (%).
Morphological characteristics
Metastatic lymph nodes (n = 150)
Non-metastatic lymph nodes (n = 115)
χ2
P value
Margin
Ill-defined
92 (61.33)
47 (40.87)
10.930
< 0.001
Well-defined
58 (38.67)
68 (59.13)
Aggregation
With
54 (36.00)
26 (22.61)
5.539
0.019
Without
96 (64.00)
89 (77.39)
Enhancement
Uneven
100 (66.67)
55 (47.83)
9.517
0.002
Uniform
50 (33.33)
60 (52.17)
Energy-spectrum parameters of metastatic and non-metastatic lymph nodes in arterial and venous phases
Metastatic lymph nodes demonstrated significantly lower IC during arterial and venous phases than non-metastatic lymph nodes (P < 0.001). Regarding the NIC, a significant reduction was observed in the arterial phase for metastatic lymph nodes in contrast to non-metastatic ones (P < 0.001). However, no significant difference in NIC in the venous phase was detected between the two groups (P > 0.05). Additionally, the λHU of metastatic lymph nodes was notably lower than that of non-metastatic lymph nodes in both arterial and venous phases (P < 0.001; Table 5).
Table 5 Energy spectrum parameters of metastatic and non-metastatic lymph nodes in arterial and venous phases, mean ± SD.
Energy spectrum parameters
Metastatic lymph nodes (n = 150)
Non-metastatic lymph nodes (n = 115)
t
P value
IC (100 μg/cm3)
Arterial phase
15.71 ± 5.92
27.00 ± 8.89
12.383
< 0.001
Venous phase
20.90 ± 8.59
33.93 ± 12.05
10.272
< 0.001
NIC
Arterial phase
0.16 ± 0.07
0.28 ± 0.11
10.810
< 0.001
Venous phase
0.44 ± 0.20
0.48 ± 0.14
1.828
0.069
λHU
Arterial phase
2.20 ± 0.68
3.41 ± 1.25
10.073
< 0.001
Venous phase
2.40 ± 1.05
3.23 ± 1.13
6.170
< 0.001
Diagnostic efficacy of GSI-related indicators via ROC analysis
The ROC analysis data indicated the diagnostic performance of various parameters for determining metastatic lymph nodes as follows. It exhibited an AUC of 0.631 [95% confidence interval (CI): 0.564-0.699, P < 0.001] for the axial short diameter, and the sensitivity stood at 83.48% and the specificity at 48.00% at a threshold of 1.28 cm. The AUC of the sagittal short diameter was measured as 0.681 (95%CI: 0.615-0.746, P < 0.001), with a sensitivity of 95.65% and a specificity of 48.67% in a threshold value of 1.14 cm. The coronal short diameter exhibited an AUC of 0.659 (95%CI: 0.592-0.725, P < 0.001), with a sensitivity of 86.09% and a specificity of 51.33% in the threshold of 1.15 cm. The IC in the arterial phase exhibited an AUC of 0.862 (95%CI: 0.817-0.908, P < 0.001), with a sensitivity of 72.17% and a specificity of 84.00% in a threshold of 21.50 × 100 μg/cm3. The AUC in the case of the venous-phase IC was 0.808 (95%CI: 0.756-0.861, P < 0.001), with sensitivity and specificity of 67.83% and 80.67%, respectively, at a threshold of 27.50 × 100 μg/cm³. The AUC of the NIC in the arterial phase was 0.831 (95%CI: 0.780-0.882, P < 0.001), with a sensitivity of 74.78% and a specificity of 80.00% in a threshold of 0.22. The arterial-phase λHU exhibited an AUC of 0.801 (95%CI: 0.742-0.860, P < 0.001), with a sensitivity of 69.57% and a specificity of 90.00% in a threshold of 2.99. Finally, the AUC of the venous-phase λHU was 0.706 (95%CI: 0.644-0.768, P < 0.001), with a sensitivity of 62.61% and a specificity of 67.33% in a threshold of 2.87. Detailed information is presented in Table 6 and Figure 2.
Figure 2 Diagnostic efficacy curves of gemstone spectral computed tomography imaging-associated indicators.
A: Diagnostic efficiency curves of the short diameters of the largest sections; B: Diagnostic efficacy curves of energy spectrum parameters in arterial and venous phases. AUC: Area under the curve; IC: Iodine concentration; NIC: Normalized iodine concentration; λHU: The slope of the spectral curve.
Table 6 Diagnostic efficacy of gemstone spectral computed tomography imaging-related indicators via receiver operating characteristic curve analysis.
Indicators
AUC
P value
95%CI
Threshold
Sensitivity
Specificity
Axial short diameter
0.631
< 0.001
0.564-0.699
1.28 cm
83.48%
48.00%
Sagittal short diameter
0.681
< 0.001
0.615-0.746
1.14 cm
95.65%
48.67%
Coronal short diameter
0.659
< 0.001
0.592-0.725
1.15 cm
86.09%
51.33%
IC in arterial phase
0.862
< 0.001
0.817-0.908
21.50 100 μg/cm3
72.17%
84.00%
IC in venous phase
0.808
< 0.001
0.756-0.861
27.50 100 μg/cm3
67.83%
80.67%
NIC in arterial phase
0.831
< 0.001
0.780-0.882
0.22
74.78%
80.00%
λHU in arterial phase
0.801
< 0.001
0.742-0.860
2.99
69.57%
90.00%
λHU in venous phase
0.706
< 0.001
0.644-0.768
2.87
62.61%
67.33%
DISCUSSION
Currently, diagnosing LNM in patients with CRC remains a challenging task for all imaging examination techniques. However, this is of great clinical significance for the accurate preoperative assessment of patients with CRC, thereby enabling optimal treatment strategy development and prognosis assessment[16].
In this research, 46.00% of the 200 patients with CRC had LNM, and 150 of the 265 Lymph nodes were metastatic. We revealed that, when it comes to diagnosing LNM in patients with CRC, GSI demonstrated an AUC of 0.897, a sensitivity of 92.59%, a specificity of 85.87%, and an accuracy of 89.50%, indicating a high potential of GSI in diagnosing LNM in patients with CRC. Furthermore, we demonstrated that metastatic lymph nodes had significantly greater short diameters in the axial, sagittal, and coronal planes than non-metastatic ones. Additionally, metastatic lymph nodes frequently demonstrated features, such as blurred margins, aggregation, and uneven enhancement, in contrast to non-metastatic lymph nodes. This is related to the fact that lymph nodes are small, bean-shaped organs distributed along lymphatic vessels. During progressive invasion from the inside out, metastatic lymph nodes generate a large number of malformed blood vessels, causing common phenomena such as blurred margins, clustered aggregation, and uneven improvement[17,18]. The IC, NIC, and λHU of metastatic lymph nodes were significantly decreased during both the arterial and venous phases (except for NIC in the venous phase, where no significant difference was observed) compared with non-metastatic lymph nodes. The mechanism underlying this difference may originate from the abnormal CRC tumor cell proliferation, causing insufficient blood supply to the medulla of lymph nodes, thereby subsequently triggering central necrosis. The numerous newly developed malformed blood vessels within metastatic lymph nodes fail to effectively increase the blood perfusion inside the lymph nodes, ultimately reducing their iodine-uptake capacity. Moreover, metastatic lymph nodes induced tissue heterogeneity; thus, their energy-spectrum curves run parallel to those of the primary lesions, whereas the energy-spectrum curves of non-metastatic lymph nodes intersect with those of the primary lesions[19-21]. The absence of a significant difference in NIC between metastatic and non-metastatic lymph nodes could be related to factors such as the partial volume effect and the limited sample size. Further in-depth investigation is warranted to validate this hypothesis. ROC analysis revealed that the AUCs of axial short diameter, sagittal short diameter, and coronal short diameter in diagnosing metastatic lymph nodes ranged from 0.600 to 0.700, indicating relatively low diagnostic efficiency of these three indexes. The AUC of the venous-phase λHU in diagnosing metastatic lymph nodes was 0.706, indicating a mediocre diagnostic performance. Conversely, AUCs of the arterial-phase IC, venous-phase IC, arterial-phase NIC, and arterial-phase λHU in diagnosing metastatic lymph nodes were 0.862, 0.808, 0.831, and 0.801, respectively, indicating that these four indicators demonstrate relatively favorable diagnostic capabilities. Notably, the arterial-phase IC boasted the highest AUC, reaching 0.862, demonstrating its superior diagnostic efficacy in identifying metastatic lymph nodes. This phenomenon may be related to the high sensitivity of the iodine-based map in quantifying iodine content. The direct measurement of iodine content renders the differences more palpable by directly reflecting the tissue’s blood supply[22,23]. Furthermore, other data relevant to diagnosing metastasis revealed that the sagittal short diameter demonstrated the highest sensitivity, accounting for 95.65%. Conversely, the arterial-phase λHU indicated the highest specificity, standing at 90.00%. Both the arterial-phase IC, with a sensitivity of 72.17% and a specificity of 84.00%, and the arterial-phase NIC, with a sensitivity of 74.78% and a specificity of 80.00%, demonstrated a commendable balance of relatively high sensitivity and specificity. Yang et al[24] revealed that the arterial-phase NIC of GSI raised the preoperative diagnostic accuracy of regional LNM in CRC to 87.1% (with the AUC increasing to 0.916), which is similar to our data. Moreover, Wu et al[25] demonstrated that GSI, when coupled with radiomics analysis of iodine-based material decomposition images, can be applied to predict the microsatellite instability status of CRC preoperatively.
Several previous investigations have assessed the diagnostic performance of various imaging modalities for detecting LNM in CRC. In particular, Kim et al[26] revealed that integrating positron emission tomography/CT with the metabolic tumor volume of the primary tumor achieved an AUC of 0.806 in predicting LNM in rectal cancer, whereas our study using GSI demonstrated superior discriminative ability (AUC = 0.897). Similarly, Rollvén et al[27] reported that conventional CT exhibited moderate predictive performance for stage III CRC LNM, with AUC values ranging from 0.596 to 0.707, substantially lower than the diagnostic accuracy observed in our analysis. Notably, Liu et al[28] revealed that magnetic resonance imaging generated an AUC of 0.890 for rectal cancer LNM detection, closely congruent with our results.
This study has several limitations. First, as a single-center retrospective study, avoiding selection bias, sample homogeneity, and the lack of an external validation cohort was difficult. Future studies need to consider incorporating multicenter samples to improve the accuracy and generalizability of the results. Second, a cost-effectiveness analysis of GSI was not conducted. Supplementing such an analysis could further clarify its clinical advantages. Third, the study did not assess long-term patient outcomes following GSI-guided treatment decisions. Future research should include such analyses to better determine its potential value.
CONCLUSION
To sum up, GSI has proven to be a valuable tool for screening LNM in patients with CRC with an AUC of 0.897, indicating remarkable diagnostic efficacy. The energy-spectrum parameters during its arterial and venous phases, namely arterial-phase IC, venous-phase IC, arterial-phase NIC, and arterial-phase λHU, also exhibit high efficiency in screening metastatic lymph nodes. Among these, the arterial-phase IC stands out with the most exceptional diagnostic performance. The results of our study provided new clues and reliable references for the screening of LNM and metastatic lymph nodes in patients with CRC and, to some extent, helped improve the efficiency of preoperative assessment for patients with CRC.
Footnotes
Provenance and peer review: Unsolicited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Gastroenterology and hepatology
Country of origin: China
Peer-review report’s classification
Scientific Quality: Grade B
Novelty: Grade C
Creativity or Innovation: Grade B
Scientific Significance: Grade C
P-Reviewer: Kobayashi S S-Editor: Bai Y L-Editor: A P-Editor: Zhang XD
Kuboki Y, Fakih M, Strickler J, Yaeger R, Masuishi T, Kim EJ, Bestvina CM, Kopetz S, Falchook GS, Langer C, Krauss J, Puri S, Cardona P, Chan E, Varrieur T, Mukundan L, Anderson A, Tran Q, Hong DS. Sotorasib with panitumumab in chemotherapy-refractory KRAS(G12C)-mutated colorectal cancer: a phase 1b trial.Nat Med. 2024;30:265-270.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 21][Cited by in RCA: 46][Article Influence: 46.0][Reference Citation Analysis (0)]
Kroon HM, Hoogervorst LA, Hanna-Rivero N, Traeger L, Dudi-Venkata NN, Bedrikovetski S, Kusters M, Chang GJ, Thomas ML, Sammour T. Systematic review and meta-analysis of long-term oncological outcomes of lateral lymph node dissection for metastatic nodes after neoadjuvant chemoradiotherapy in rectal cancer.Eur J Surg Oncol. 2022;48:1475-1482.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 3][Cited by in RCA: 20][Article Influence: 6.7][Reference Citation Analysis (0)]
