Skip to main content
Insights into Imaging
Download PDF
  • Original Article
  • Open access
  • Published:

Performance of dual-layer spectrum CT virtual monoenergetic images to assess early rectal adenocarcinoma T-stage: comparison with MR

Insights into Imaging volume 15, Article number: 11 (2024) Cite this article

Abstract

Objectives

To evaluate the image quality and utility of virtual monoenergetic images (VMI) of dual-layer spectrum computed tomography (DLSCT) in assessing preoperative T-stage for early rectal adenocarcinoma (ERA).

Methods

This retrospective study included 67 ERA patients (mean age 62 ± 11.1 years) who underwent DLSCT and MR examination. VMI 40–200 keV and poly energetic image (PEI) were reconstructed. The image noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and tumor contrast of different energy levels were calculated and compared, respectively. Two radiologists independently assess the image quality of the VMIs and PEI using 5-point scales. The diagnostic accuracies of DLSCT and HR-MRI for ERA T-staging were evaluated and compared.

Results

The maximum noise was observed at VMI 40 keV, and noise at VMI 40–200 keV in the arterial and venous phases showed no significant difference (all p > 0.05). The highest SNR and CNR were obtained at VMI 40 keV, significantly greater than other energy levels and PEI (all p < 0.05). Tumor contrast was more evident than PEI at 40–100 keV in the arterial phase and at 40 keV in the venous phase (all p < 0.05). When compared with PEI, VMI 40 keV yielded the highest scores for overall image quality, tumor visibility, and tumor margin delineation, especially in the venous phase (p < 0.05). The overall diagnostic accuracy of DLSCT and HR-MRI for T-stage was 65.67 and 71.64% and showed no significant difference (p > 0.05).

Conclusions

VMI 40 keV improves image quality and accuracy in identifying lesions, providing better diagnostic information for ERA staging.

Critical relevance statement

Low-keV VMI from DLSCT can improve tumor staging accuracy for early rectal carcinoma, helping guide surgical intervention decisions, and has shed new light on the potential breakthroughs of assessing preoperative T-stage in RC.

Keypoints

• Compared with PEI, low-keV VIM derived from DLSCT, particularly at the 40 keV, significantly enhanced the objective and subjective image quality of ERA.

• Using VMI 40 keV helped increase lesion detectability, leading to improved diagnostic accuracy for ERA.

• Low-keV VMI from DLSCT has shed new light on the potential breakthroughs of assessing preoperative T-stage in RC.

Graphical Abstract

Introduction

Colorectal cancer (CRC) is the third most prevalent form of cancer and the third most common cause of cancer-related deaths [1]. Rectal cancer (RC) accounts for one-third of all CRC cases. Early rectal cancer (ERC) is categorized as stage I, where the tumor grows through the muscularis mucosa into the submucosa (T1) or muscularis propria (T2) without the involvement of the lymph nodes (N0) or distant metastasis (M0) [2]. The relative incidence of ERC has increased since the implementation of gastrointestinal screening programs [1]. Although total mesorectal excision (TME) remains the standard procedure for most rectal cancer patients, the rising occurrence of early-stage cancer has emphasized the significance of local excision (LE) and organ preservation as viable treatment options for these patients [3, 4]. These options can be pursued without compromising anorectal function. Therefore, precise assessment of the T-staging is crucial in informing therapeutic decision-making for RC patients.

Currently, several imaging modalities are available for evaluating the preoperative T-staging of RC, including endorectal ultrasonography (ERUS) and high-resolution magnetic resonance imaging (HR-MRI). ERUS is effective for evaluating early-stage of low and middle-rectal cancer [5]. However, its performance for evaluating intramural T2 lesions is unsatisfactory, which are often overstage with a low sensitivity (30%) [6,7,8]. Furthermore, ERUS is operator dependent and not applicable in stenosing lesions. HR-MRI is recommended by the National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines due to its superior soft tissue visualization [9]. Evaluation of mrT-staging is mainly based on the depth of tumor penetration into the rectal wall and extramural spread into the mesorectum and adjacent structures, which is largely performed with T2-weighted imaging. Nevertheless, the limitation of HR-MRI is its unreliable distinction between T1 and T2 tumors and the overstaging of T2 tumors as T3 tumors [10, 11]. Sensitivity in MRI resulted 58% in T1 and 30% in T2 cases, respectively [8]. Meanwhile, this strategy is associated with a longer waiting time and higher expenditure. Although computed tomography (CT) cannot replace ERUS and HR-MRI in T category, it has an important role in evaluating distant metastases. Therefore, CT is still advocated as a routine diagnostic modality to evaluate RC staging before treatment, stage recurrent disease, and postoperative monitoring by many current guidelines [12].

Dual-energy CT (DECT) provides quantitative information about tissue composition and improves lesion visualization. For instance, the visualization of abnormal enhanced bowel wall (ischemic small bowel and inflammatory bowel disease) has been improved by using low-keV images and iodine maps of DECT [13,14,15,16,17]. Studies have also demonstrated DECT is effective in identifying low-enhanced abnormalities like hypovascular liver lesions and pancreatic adenocarcinoma [18,19,20,21]. However, some limitations remain, such as artifacts caused by cross-scattered radiation in dual-source CT, misregistration between high and low voltage data, and fast kV switching waveform resembles a size wave rather than a squared wave, which leads to poor spectrum separation [22, 23]. Dual-layer spectral detector CT (DLSCT), which uses two detector layers and an X-ray tube, has recently emerged as a novel and promising imaging modality for radiological evaluation [24]. Compared to dual-energy or fast kilovoltage switching techniques, the benefit of DLSCT is that the two measurement projection data are perfectly matched in temporal and spatial co-registration, and energy spectrum scanning is realized simultaneously as conventional scanning without any additional steps or radiation agents [24]. DLSCT can also provide quantitative information, reduce photon noise and beam-hardening effects, and improve CT image quality and lesion visualization. Our previous results demonstrated that quantitative parameters derived from DLSCT could be used to assess the pT stage and histological differentiation in patients with colorectal adenocarcinoma [25]. Taguchi et al. showed that virtual monochromatic imaging (VMI) could significantly improve fecal-tagged CT colonography image quality and enhance electronic cleansing performance when using DLSCT [26]. However, there is currently little research on the clinical utility of DLSCT for evaluating the tumor stage of early rectal adenocarcinoma. We hypothesized that DLSCT VMI could improve the accuracy of local tumor staging for ERC by increasing tissue differentiation based on material decomposition. Therefore, the purpose of this study was twofold. Firstly, to investigate and compare image quality of DLSCT-derived VMI with PEI in the patients with ERA. Secondly, to evaluate DLSCT VMI would improve diagnostic accuracy of T-staging compared to HR-MRI.

Materials and methods

Participants

Between May 2021 and March 2023, 82 patients with pathologically confirmed early rectal adenocarcinoma (ERA) were retrospectively enrolled in this study. The review board of our hospital approved the study design. The inclusion criteria were as follows: (1) histopathologically confirmed ERA, including pT and pN stage and histological grade; (2) complete clinical information; and (3) all patients underwent DLSCT and HR-MRI examination within a week before surgery with time interval between MRI/CT examination and surgery of < 2 weeks. The exclusion criteria were as follows: (1) severe allergy to the iodinated contrast agent, impaired kidney function (creatinine clearance less than 40 ml/min/1.73 m2); and (2) poor image quality or severe artifacts on MR images.

Imaging protocol

CT scanning was performed using a IQon Spectral CT (Philips Healthcare, Best, The Netherlands). The CT scan range was from the diaphragm to the symphysis pubis. DLSCT scan parameters were the following: tube voltage, 120 kV; tube current, automated exposure modulation; detection collimation, 64 × 0.625 mm; rotation time, 0.4 s; and helical pitch, 1.1. For contrast-enhanced CT scans, a body-weight-adapted dose of nonionic iodine contrast agent (ultrafast 370, Bayer Healthcare, 1.5 ml/kg body weight) was administered via the antecubital vein with a constant flow of 2.5 ml/s. Bolus tracking was performed using a region of interest within the abdominal aorta, and arterial phase (AP) acquisition was started when the attenuation reached a predefined threshold of 150 Hounsfield units (HU). Venous-phase (VP) images were acquired 40 s after AP.

Conventional and quantitative spectral images were reconstructed and analyzed using post-processing software (IntelliSpace Portal Version 10.0, Philips Healthcare). A VMI of 40–200 keV was reconstructed using a reconstruction algorithm (Spectral Level 3, Philips Healthcare) with a 10-keV interval. The PEI was reconstructed using an iterative reconstruction algorithm (iDose4, Philips Healthcare) with a constant standard soft tissue kernel. All images were reconstructed at a slice thickness of 2 mm for further analysis.

The patient examination protocol for the CT scanner included the mean mAs and dose index of the entire examination. These values were documented to ensure accurate and thorough recording.

Rectal MR was conducted using various 3.0 T MR scanners from Siemens Healthcare (Prisma, Verio) and Philips Healthcare (Ingenia). MR sequences included T1WI, high-resolution T2WI, T1-weighted enhanced images (with Magnevist®, Bayer HealthCare, Guangzhou, China, at a dose of 0.2 ml/kg of body weight), and conventional DWI (b = 0, 1000 s/mm2). The detailed parameters of the MR protocol are listed in Table S1.

DLSCT image objective analysis

Two radiologists (with 25 and 16 years of experience in gastrointestinal radiology), blinded to the clinical and pathological information, independently evaluated the DLSCT data and performed an objective analysis.

Regions of interest (ROIs) were drawn manually to measure the mean attenuation of the tumor (HU tumor) and normal intestinal wall tissue (HU intestinal wall) on –40–200 keV VMI and PEI images in the arterial and venous phases. The ROI was drawn by selecting identical axial sections where the maximum diameter of the lesion was located and covering the lesion as much as possible, avoiding fat and necrosis. Each ROI was kept in a consistent position and size on all sets of VMIs and PEI in both the arterial and venous phases and measured twice to ensure consistency with the averaged values used for the analysis. The normal intestinal wall was delineated into the upper and lower layers adjacent to the tumor. Image noise was defined as the standard deviation within the muscle (SD muscle) in the same slice of 10–15 mm2 to eliminate selective disparities. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of rectal tumors and tumor contrast were calculated using the following formulas:

$$\textrm{SNR}={\textrm{HU}}_{\textrm{tumor}}/{\textrm{SD}}_{\textrm{muscle}}$$
$$\textrm{CNR}=\left({\textrm{HU}}_{\textrm{tumor}}-{\textrm{HU}}_{\textrm{intestinal}\ \textrm{wall}}\right)/{\textrm{SD}}_{\textrm{muscle}}$$
$$\textrm{Tumor}\ \textrm{contrast}={\textrm{HU}}_{\textrm{tumor}}-{\textrm{HU}}_{\textrm{bowel}\ \textrm{wall}}$$

DLSCT image subjective analysis

The aforementioned two radiologists reviewed the DLSCT VMI images. Radiologists used 5-point scales to assess the quality of the images, tumor visibility, and tumor margin delineation in a blinded and randomized manner. The scale ranged from 1 (unacceptable) to 5 (excellent) for overall image quality and 1 (unacceptable) to 5 (excellent) for tumor visibility and margin delineation.

DLSCT and MRI evaluation in T-staging of ERC

All DLSCT VMIs and HR-MRI examinations were analyzed in consensus by the aforementioned two radiologists, respectively. Both of them were blind to the histopathologic examinations.

Tumor staging based on the spectrum CT and MR was separately analyzed according to the AJCC (American Joint Committee on Cancer), 8th edition. To minimize recall bias, the DLSCT and HR-MRI evaluation were conducted as two separate assignments, with a 3-month washout period in between. DLSCT T-staging was determined using objective and subjective analysis results that provide the best tumor visibility. MR T-stage evaluation was conducted based on the multiparameter MRI, including T1WI, T2WI, and DWI sequences [27].

Histological analysis

All patients underwent surgery within 1 week after DLSCT and HR-MRI examinations. The tumor specimens were reviewed by a pathologist with 14 years of experience in gastroenterology using HE and immunohistochemical staining. TNM stages and tumor grades were determined based on AJCC, 8th edition. Early rectal tumors are T1 or T2 tumors without lymph nodes or distant metastasis (T1-2, N0, M0). T1: tumor invades the submucosa (through the muscularis mucosa, but not into the muscularis propria). T2: tumor invades the muscularis propria. Tumor grade was classified according to the percentage of gland formation, with grade 1 indicating well-differentiated tumors (> 95% gland formation), grade 2 indicating moderately differentiated tumors (50–95% gland formation), and grade 3 indicating poorly differentiated tumors (< 50% gland formation).

Statistical analysis

Statistical analyses were performed using SPSS (version 20.0; SPSS Inc., Chicago, IL, USA) and GraphPad Prism 9.0 (GraphPad Software, San Diego, CA, USA).

The inter-reader reproducibility of the noise, HU of the lesion, and the intestinal wall was calculated using the intraclass correlation coefficient (ICC). The Kolmogorov–Smirnov test was used to assess data distribution. Continuous variables are presented as mean (standard deviation) or median (interquartile range) depending on the data distribution. The nonparametric Friedman test was used to compare quantitative variables between VMI and PEI in the arterial and venous phases. The kappa coefficient (k) was used to measure the inter-reader agreement of the overall image quality, tumor visibility, and delineation (k = 0.8–1.00 excellent; 0.61–0.8 substantial; 0.41–0.60 moderate; 0.21–0.40 fair; 0.00–0.20 poor). Subjective quality scores were compared using nonparametric tests (Kruskal–Wallis test, followed by Steel–Dwass post hoc test).

Using the pathological results as a reference, the overall accuracy, overstaging, and under-staging rates of DLSCT and MR were calculated. The chi-square test was used to compare the diagnostic performances of DLSCT and HR-MRI. A p-value less than 0.05 was considered statistically.

Results

Participants

In total, 67 pathologically confirmed early rectal adenocarcinoma patients were enrolled in the study, consisting of 25 females and 42 males (mean age, 62 ± 11.1 years) (Fig. 1). Detailed patient demographics are shown in Table 1.

Fig. 1
figure 1

Flowchart shows patient selection

Full size image
Table 1 Demographic characteristics of the patients
Full size table

Objective image quality

Inter-observer agreement was excellent (intraclass coefficient > 0.90) for all the variables. The ICCs ranged from 0.900 to 0.995 for HU tumor, HU intestinal wall, and SD muscle of VMI 40–200 keV and PEI in the arterial and venous phases. Details of the ICCs values are provided in Table S2.

The noise levels in the arterial and venous phases were low at all VMI energy levels. The maximum noise was observed at 40 keV [AP 11.05 (9.70, 12.95); VP 11.10 (10.00, 12.55)], followed by a slight decrease with increased energy levels, and the noise was relatively stable at all the energy levels. Multiple comparisons demonstrated that the noise at 50–200 keV was significantly lower than that of PEI in the AP (all p < 0.05) and that the noise at all energy levels was significantly lower than that of PEI in the VP (all p < 0.05). Noise at 40–200 keV in the arterial and venous phases showed no significant differences (all p > 0.99) (Fig. 2a).

Fig. 2
figure 2

Objective image characteristics for VMI from 40 to 200 keV and PEI. The data for image noise (a), tumor SNR (b), tumor CNR (c), and tumor contrast (d) at VMI 40–200 keV and PEI in the arterial and venous phases are presented using box-and-whisker plots. Each box in the plot shows the upper and lower quartiles of the data, while the median value is displayed as a horizontal line

Full size image

The tumor SNR and CNR in the arterial and venous phases are shown in Table 2. As the energy level decreased, the SNR and CNR gradually increased. The VMI 40 keV provided the highest tumor SNR [15.56 (13.93, 18.00)] in the AP, although there was no statistical difference between 40 and 50 keV [10.48 (8.43, 11.97)] (p > 0.99, Fig. 2a). Tumor SNR of VMI 40–60 keV was significantly higher than that of PEI [15.56 (13.93, 18.00)–8.70 (6.76, 9.78) vs. 6.00 (5.31, 7.82), p < 0.05], while tumor SNR of VMI 130–200 keV was significantly lower than that of PEI [5.26 (3.91, 6.30)–4.68 (3.52, 5.87), p < 0.05]. In the VP, 40 keV achieved the maximum tumor SNR [19.21 (16.25, 23.00)], while tumor SNR at the low energy levels showed no difference between 40, 50, and 60 keV (all p > 0.99) (Fig. 2b).

Table 2 Subjective image quality analysis
Full size table

The tumor CNR ranged from [10.71 (8.34, 13.00), 40 keV] to [2.53 (0.86, 2.91), 200 keV] and from [12.00 (8.62, 15.69), 40 keV] to [3.77 (2.39, 4.74), 200 keV] for the arterial and venous phases, respectively. The highest CNR was obtained at 40 keV [10.71 (8.34, 13.00), AP; 12.00 (8.62, 15.69), VP], which was significantly greater than that at the other energy levels and PEI (all p < 0.05) (Fig. 2c).

Tumor contrast gradually increased as the monoenergetic level decreased in both the arterial and venous phases. The maximum tumor contrast in the AP was obtained at 40 keV, but it did not show a significant difference when compared to that at 50 and 60 keV (Figs. 2d and 3). In the VP, the tumor contrast at 40 keV reached the maximum value, which was significantly higher than that at the other energy levels and PEI (all p < 0.05) (Figs. 2d and 4). The detailed data of objective image quality analysis was shown in Table S3.

Fig. 3
figure 3

Contrast-enhanced axial arterial phase in 66-year-old female with T2 stage rectal adenocarcinoma. VMI 40–200 keV of arterial phase shows that tumor SNR, tumor CNR, and tumor contrast increases as the energy level decreases. The highest SNR, CNR, and tumor contrast were obtained on the VMI 40 keV, followed by 50 and 60 keV. There was no difference in the SNR, tumor contrast among VMI 40, 50, and 60 keV. The best image quality, tumor visibility, and tumor delineation were obtained at VMI 40 keV

Full size image
Fig. 4
figure 4

Contrast-enhanced axial venous phase in the same patient in Fig. 3. VMI 40–60 keV outperformed PEI in objective analysis and subjective scores. The highest SNR, CNR, and tumor contrast were obtained on the VMI 40 keV, which were significantly higher than other energy levels and PEI (all p < 0.05). VMI 40 keV in the venous phase yielded the highest overall image quality, tumor visibility, and tumor margin delineation scores

Full size image

Subjective image quality

Based on the objective analysis results, a subjective analysis was performed using VMI 40–60 keV and PEI. The agreement between the overall image quality, tumor visibility, and tumor margin delineation ranged from substantial to excellent (kappa value = 0.688, 0.897, and 0.763 in the AP and 0.688, 0.897, and 0.763 in the VP), indicating good inter-observer reliability. VMI 40–60 keV outperformed PEI in all criteria as per the subjective scores. VMI of 40 keV yielded the highest scores for overall image quality, tumor visibility, and tumor margin delineation, especially in the VP (p < 0.001). Table 3 summarizes the results of subjective analyses.

Table 3 Performance of DLSCT and MRI for staging pT1 and pT2 rectal cancers
Full size table

Comparison of diagnostic accuracy for T-staging between DLSCT and HR-MRI

According to the objective and subjective analysis results, a VMI 40–60 keV of arterial and venous phases was selected to assess the T-stage. The diagnostic accuracies of DLSCT and HR-MRI for T-staging were evaluated and compared. Inter-observer agreement in DLSCT and MRI T-stage was significant (DLSCT: kappa value = 0. 82, p < 0.001; MRI: kappa value = 0.75, p < 0.001). For ERA T-stage, the overall DLSCT diagnostic accuracy was 65.67% (44/67, 57.28% sensitivity and 65.77% specificity, 95% confidence interval 54.3 to 77.1%), with six pT2 patients being underdiagnosed with T1, eight pT1 patients overdiagnosed with T2, and nine pT2 patients overdiagnosed with T3. The diagnostic accuracy of HR-MRI was 71.64% (48/67, 66.31% sensitivity and 73.86% specificity, 95% confidence interval 60.9 to 82.4%), including five pT2 patients mistaken as T1, six pT1 patients misdiagnosed as T2, and eight pT2 patients inaccurately diagnosed as T3. The accuracy for T1-2 (χ2 = 0.555, p = 0.45), T1 (χ2 = 0.337, p = 0.56), and T2 staging (χ2 = 0.337, p = 0.56) between DLSCT and HR-MRI showed no significant differences (all p > 0.05). The detailed results are presented in Table 3. The representative cases are shown in Figs. 5 and 6.

Fig. 5
figure 5

A 59-year-old male with T1 rectal adenocarcinoma who underwent dual-layer spectral CT and HR-MRI. VMI 40–60 keV in the arterial phase (a–c) and venous phase (d,e) demonstrate that the tumor is confined to the submucosa. The VMI 40 keV of venous phase (d) provides the best tumor conspicuity and margin delineation and exhibits no significant difference compared to MRI (g: axial T2WI; h: Gd-DPTA enhanced T1WI; i: DWI at b = 1000 s/mmm2)

Full size image
Fig. 6
figure 6

A 76-year-old female with T2 rectal adenocarcinoma who underwent dual-layer spectral CT and HR-MRI. VMI 40–60 keV in the arterial phase (a–c) and venous phase (d,e) demonstrate that the tumor has grown into the muscularis propria and has not yet broken through the serosal layer. The VMI 40 keV of venous phase (d) provides the same capability of stage evaluation as MRI (g: axial T2WI; H: Gd-DPTA enhanced T1WI; i: DWI at b = 1000 s/mmm2)

Full size image

Discussion

In the current study, we evaluated the imaging quality of VMI using DLSCT in ERA and compared its T-stage evaluation capability to that of HR-MRI. Our findings indicate that when compared with PEI, low noise levels were obtained from VMI 40 to 200 keV in both the arterial and venous phases. The best image quality and lesion detectability were achieved with a VMI of 40 keV in the VP, which showed higher SNR, higher CNR, better tumor contrast, and acceptable image noise. In addition, we found that the accuracy of DLSCT was comparable to that of HR-MRI in evaluating the T-stage of ERA.

Monochromatic images are generally synthesized through basis material decomposition (projection or image domain) and known densities of the basis materials [28]. Image noise is an important criterion for evaluating image quality. Our study showed that the highest noise level was observed at 40 keV and that the noise magnitude decreased as the VMI energy increased. In the venous phase, the noise at all VMI energy levels was significantly lower than that of the PEI. This finding is consistent with the results of a previous study by Kalisz et al., who found that the VMI obtained from DLSCT had low noise across the entire energy spectra [29]. Although previous phantom studies and patient cohorts have shown high image noise in low-energy VMI [30], which limits its utility and diagnostic capabilities, we found no significant noise differences across the entire scope of VMIs, regardless of whether it was in the arterial or venous phase. This result is similar to that of a study on abdominal phantoms by Sellerer et al., which showed that DLSCT produced no significant alteration in noise throughout the entire range of VMIs [31]. This can be attributed to the perfect alignment of the low- and high-energy spectra in the projection domain, which reduces beam-hardening artifacts and anti-correlated noise [32].

We found that the tumor SNR, tumor CNR, and tumor contrast gradually increased as the energy levels decreased, and the highest tumor contrast was obtained at VMI 40 keV with a significant improvement. VMI at 40–60 keV had a higher SNR than PEI, with VMI at 40 keV being the highest. This is likely owing to the following factors. First, the higher photoelectric attenuation of the low-energy VMI image was close to the K-edge of iodine (33.17 keV), resulting in higher absorption and enhanced iodine contrast in the image [20]. Second, DLSCT provides a constant low noise across the entire spectrum of energies. Previous studies have shown that VMIs obtained from detector-based spectral CT achieved significant SNR and CNR improvements compared to conventional 120 kVp images. For instance, Lee et al. showed that VMI40 provided optimal CNR for abnormal and normal small bowel walls on dual-layer dual-energy CT enterography (DE-CTE) [17]. Nagayama et al. demonstrated that low-keV VMI improved the image quality and yielded adequate diagnostic tumor detectability in patients with dual pancreatic adenocarcinoma and hypovascular hepatic metastases [20]. Concordant with these findings, our subjective analysis revealed that the low VMI series performed better than did the PEI series with superior tumor quality, tumor visibility, and more prominent tumor delineation, while a VMI of 40 keV was considered optimal for improved lesion delineation and detection performance and increased readers’ confidence. Additionally, we found that the SNR, CNR, and tumor contrast in the venous phase were higher than those in the AP. This may be due to the immature neovascularization of ERC, which makes the contrast agent more susceptible to extravasation. The contrast agent is easily retained and distributed in the venous phase; therefore, it can better reflect the attenuation of iodine in tumor tissue [33].

Surgical treatment strategy for rectal cancer depends on accurate preoperative staging. Local excision is suitable only for early-stage patients with cT1N0 and low grade. For patients with cT1/T2 and negative N status, transabdominal radical resection without neoadjuvant therapy is recommended [4, 34]. MR is the preferred modality for rectal cancer staging because of its ability to evaluate the depth of tumor invasion. However, distinguishing between T1 and T2 lesions remains challenging. A recent Dutch population-based study demonstrated that 54.7% of patients with pT1 tumors were overstaged by MRI alone [35]. In our study, the depth of tumor invasion into the rectal bowel wall was accurately evaluated by using the combination of the high-contrast of VMI 40 keV and noticeable tumor enhancement. DLSCT provided 65.67% diagnostic accuracy for the T category, similar to that of MRI, and a shorter scan time could reduce the possibility of motion artifacts.

Our study had several limitations. First, it was a retrospective study with a small sample size. A prospective study with larger samples is required to verify our results. Second, we only focused on the feasibility of the VMI for the evaluation of early rectal adenocarcinoma. Other parameters obtained from dual-energy CT, such as iodine map, can improve the visibility of bowel wall hypoenhancement [14, 36]. Additionally, color-coded effective Z-images after enhancement were helpful in increasing the tumor conspicuity. Third, DLSCT image quality is affected by other parameters such as the reconstruction kernel, collimation, slice thickness, iodine contrast material dose, patient size, and dose level [37,38,39]. Notably, low-energy VMI can improve image contrast with lower amounts of iodine contrast agents and radiation doses. Further studies should focus on improving image quality and lesion detectability with reduced radiation dose and contrast optimization. Last, our study focused on the diagnostic capability of VMI in evaluating T-stage of early rectal cancer. Besides VMI, dual-energy CT can provide multiple quantitative parameters regarding tissue’s biological information and offers a non-invasive method for predicting histopathological and molecular identification [40,41,42,43]. Further research should be carried out to explore the relationship between dual-energy CT-derived parameters and genotype classification of rectal cancer.

Conclusion

In conclusion, low-keV VIM derived from DLSCT improved early rectal adenocarcinomas’ objective and subjective image quality. VMI 40 keV of DLSCT provided low image noise and high SNR, CNR, and tumor contrast, which improved the accuracy of lesion identification and provided better diagnostic information for the staging of ERA. The findings provide evidence that DLSCT has shed new light on the potential breakthroughs of assessing preoperative T-stage in RC.

Availability of data and materials

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Abbreviations

CNR:

Contrast-to-noise ratio

DECT:

Dual-energy CT

DLSCT:

Dual-layer spectrum computed tomography

ERA:

Early rectal adenocarcinoma

ERUS:

Endorectal ultrasonography

HR-MRI:

High-resolution magnetic resonance imaging

NCCN:

National Comprehensive Cancer Network

PEI:

Poly energetic image

RC:

Rectal carcinoma

SNR:

Signal-to-noise ratio

TME:

Total mesorectal excision

VMI:

Virtual monoenergetic images

References

  1. Siegel RL, Wagle NS, Cercek A, Smith RA, Jemal A (2023) Colorectal cancer statistics, 2023. CA Cancer J Clinic 73:233–254

    Article  Google Scholar 

  2. Lezoche E, Baldarelli M, De Sanctis A, Lezoche G, Guerrieri M (2007) Early rectal cancer: definition and management. Dig Dis 25:76–79

    Article  PubMed  Google Scholar 

  3. Smits LJ, van Lieshout AS, Grüter AA, Horsthuis K, Tuynman JB (2022) Multidisciplinary management of early rectal cancer–The role of surgical local excision in current and future clinical practice. Surg Oncol 40:101687

    Article  PubMed  Google Scholar 

  4. Glynne-Jones R, Wyrwicz L, Tiret E et al (2017) Rectal cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 28:iv22–iv40

    Article  CAS  PubMed  Google Scholar 

  5. Puli SR, Bechtold ML, Reddy JB, Choudhary A, Antillon MR, Brugge WR (2009) How good is endoscopic ultrasound in differentiating various T stages of rectal cancer? Meta-analysis and systematic review. Ann Surg Oncol 16:254–265

    Article  PubMed  Google Scholar 

  6. Bartram C, Brown G (2002) Endorectal ultrasound and magnetic resonance imaging in rectal cancer staging. Gastroenterol Clin 31:827–839

    Article  Google Scholar 

  7. Schaffzin DM, Wong WD (2004) Endorectal ultrasound in the preoperative evaluation of rectal cancer. Clin Colorectal Cancer 4:124–132

    Article  PubMed  Google Scholar 

  8. Oien K, Forsmo HM, Rosler C, Nylund K, Waage JE, Pfeffer F (2019) Endorectal ultrasound and magnetic resonance imaging for staging of early rectal cancers: how well does it work in practice? Acta Oncol 58:S49–S54

    Article  PubMed  Google Scholar 

  9. Iafrate F, Laghi A, Paolantonio P et al (2006) Preoperative staging of rectal cancer with MR Imaging: correlation with surgical and histopathologic findings. Radiographics 26:701–714

    Article  PubMed  Google Scholar 

  10. Beets-Tan RGH, Beets GL (2004) Rectal Cancer: Review with Emphasis on MR Imaging. Radiology 232:335–346

    Article  PubMed  Google Scholar 

  11. Kaur H, Choi H, You YN et al (2012) MR imaging for preoperative evaluation of primary rectal cancer: practical considerations. Radiographics 32:389–409

    Article  PubMed  Google Scholar 

  12. Raman SP, Chen Y, Fishman EK (2015) Evolution of imaging in rectal cancer: multimodality imaging with MDCT, MRI, and PET. J Gastroint Oncol 6:172–184

    Google Scholar 

  13. Potretzke TA, Brace CL, Lubner MG, Sampson LA, Willey BJ, Lee FT Jr (2015) Early small-bowel ischemia: dual-energy CT improves conspicuity compared with conventional CT in a swine model. Radiology 275:119–126

    Article  PubMed  Google Scholar 

  14. Fulwadhva UP, Wortman JR, Sodickson AD (2016) Use of dual-energy CT and iodine maps in evaluation of bowel disease. Radiographics 36:393–406

    Article  PubMed  Google Scholar 

  15. Dane B, Sarkar S, Nazarian M et al (2021) Crohn disease active inflammation assessment with iodine density from dual-energy CT enterography: comparison with histopathologic analysis. Radiology 301:144–151

    Article  PubMed  Google Scholar 

  16. Lourenco PD, Rawski R, Mohammed MF, Khosa F, Nicolaou S, McLaughlin P (2018) Dual-energy CT iodine mapping and 40-keV monoenergetic applications in the diagnosis of acute bowel ischemia. AJR Am J Roentgenol 211:564–570

    Article  PubMed  Google Scholar 

  17. Lee SM, Kim SH, Ahn SJ, Kang H-J, Kang JH, Han JK (2018) Virtual monoenergetic dual-layer, dual-energy CT enterography: optimization of keV settings and its added value for Crohn’s disease. Eur Radiol 28:2525–2534

    Article  PubMed  Google Scholar 

  18. Macari M, Spieler B, Kim D et al (2010) Dual-source dual-energy MDCT of pancreatic adenocarcinoma: initial observations with data generated at 80 kVp and at simulated weighted-average 120 kVp. AJR Am J Roentgenol 194:W27–W32

    Article  PubMed  Google Scholar 

  19. Gupta S, Wagner-Bartak N, Jensen CT et al (2016) Dual-energy CT of pancreatic adenocarcinoma: reproducibility of primary tumor measurements and assessment of tumor conspicuity and margin sharpness. Abdom Radiol (NY) 41:1317–1324

    Article  PubMed  Google Scholar 

  20. Nagayama Y, Iyama A, Oda S et al (2019) Dual-layer dual-energy computed tomography for the assessment of hypovascular hepatic metastases: impact of closing k-edge on image quality and lesion detectability. Eur Radiol 29:2837–2847

    Article  PubMed  Google Scholar 

  21. Sudarski S, Apfaltrer P, Nance JW Jr et al (2014) Objective and subjective image quality of liver parenchyma and hepatic metastases with virtual monoenergetic dual-source dual-energy CT reconstructions: an analysis in patients with gastrointestinal stromal tumor. Acad Radiol 21:514–522

    Article  PubMed  Google Scholar 

  22. Krauss B, Schmidt B, Flohr TG (2011) Dual source CT. In: Dual Energy CT in clinical practice. Springer, pp 11–20

    Chapter  Google Scholar 

  23. Zou Y, Silver MD (2008) Analysis of fast kV-switching in dual energy CT using a pre-reconstruction decomposition technique. In: Medical imaging 2008: physics of medical imaging. SPIE, pp 392–403

    Google Scholar 

  24. Rassouli N, Etesami M, Dhanantwari A, Rajiah P (2017) Detector-based spectral CT with a novel dual-layer technology: principles and applications. Insights Imaging 8:589–598

    Article  PubMed  PubMed Central  Google Scholar 

  25. Chen W, Ye Y, Zhang D et al (2022) Utility of dual-layer spectral-detector CT imaging for predicting pathological tumor stages and histologic grades of colorectal adenocarcinoma. Front Oncol 12:1002592

    Article  PubMed  PubMed Central  Google Scholar 

  26. Taguchi N, Oda S, Imuta M et al (2018) Dual-energy computed tomography colonography using dual-layer spectral detector computed tomography: Utility of virtual monochromatic imaging for electronic cleansing. Eur J Radiol 108:7–12

    Article  PubMed  Google Scholar 

  27. Nougaret S, Reinhold C, Mikhael HW, Rouanet P, Bibeau F, Brown G (2013) The use of MR imaging in treatment planning for patients with rectal carcinoma: have you checked the "DISTANCE"? Radiology 268:330–344

    Article  PubMed  Google Scholar 

  28. Yu L, Leng S, McCollough CH (2012) Dual-energy CT–based monochromatic imaging. AJR Am J Roentgenol 199:S9–S15

    Article  PubMed  Google Scholar 

  29. Kalisz K, Rassouli N, Dhanantwari A, Jordan D, Rajiah P (2018) Noise characteristics of virtual monoenergetic images from a novel detector-based spectral CT scanner. Eur J Radiol 98:118–125

    Article  PubMed  Google Scholar 

  30. Yu L, Christner JA, Leng S, Wang J, Fletcher JG, McCollough CH (2011) Virtual monochromatic imaging in dual-source dual-energy CT: radiation dose and image quality. Med Phys 38:6371–6379

    Article  PubMed  PubMed Central  Google Scholar 

  31. Sellerer T, Noël PB, Patino M et al (2018) Dual-energy CT: a phantom comparison of different platforms for abdominal imaging. Eur Radiol 28:2745–2755

    Article  PubMed  Google Scholar 

  32. Altman A, Kafri G, Shenhav S (2022) Dual-Energy: The Philips ApproachSpectral Imaging: Dual-Energy, Multi-Energy and Photon-Counting CT. Springer, pp 29–44

    Book  Google Scholar 

  33. Saclarides TJ, Speziale NJ, Drab E, Szeluga DJ, Rubin DB (1994) Tumor angiogenesis and rectal carcinoma. Dis Colon Rectum 37:921–926

    Article  CAS  PubMed  Google Scholar 

  34. Benson AB, Venook AP, Al-Hawary MM et al (2022) Rectal cancer, version 2.2022, NCCN clinical practice guidelines in oncology. J Natl Compr Cancer Netw 20:1139–1167

    Article  CAS  Google Scholar 

  35. Detering R, Oostendorp S, Meyer V et al (2020) MRI cT1–2 rectal cancer staging accuracy: a population-based study. J British Surg 107:1372–1382

    Article  CAS  Google Scholar 

  36. Obmann MM, Punjabi G, Obmann VC et al (2022) Dual-energy CT of acute bowel ischemia. Abdom Radiol (NY) 47:1660–1683

    Article  PubMed  Google Scholar 

  37. Morsbach F, Zhang Y-H, Martin L, Lindqvist C, Brismar T (2019) Body composition evaluation with computed tomography: Contrast media and slice thickness cause methodological errors. Nutrition 59:50–55

    Article  PubMed  Google Scholar 

  38. Primak AN, Giraldo JCR, Eusemann CD et al (2010) Dual-source dual-energy CT with additional tin filtration: Dose and image quality evaluation in phantoms and in-vivo. AJR Am J Roentgenol 195:1164

    Article  PubMed  PubMed Central  Google Scholar 

  39. Jiang X, Yang X, Hintenlang DE, White RD (2021) Effects of patient size and radiation dose on iodine quantification in dual-source dual-energy CT. Acad Radiol 28:96–105

    Article  PubMed  Google Scholar 

  40. Yanagawa M, Morii E, Hata A et al (2016) Dual-energy dynamic CT of lung adenocarcinoma: correlation of iodine uptake with tumor gene expression. Eur J Radiol 85:1407–1413

    Article  PubMed  Google Scholar 

  41. Wang X, Liu D, Zeng X et al (2021) Dual-energy CT quantitative parameters for evaluating Immunohistochemical biomarkers of invasive breast cancer. Cancer Imaging 21:1–10

    Article  Google Scholar 

  42. Wu J, Lv Y, Wang N et al (2019) The value of single-source dual-energy CT imaging for discriminating microsatellite instability from microsatellite stability human colorectal cancer. Eur Radiol 29:3782–3790

    Article  PubMed  Google Scholar 

  43. Mao LT, Chen WC, Lu JY et al (2023) Quantitative parameters in novel spectral computed tomography: Assessment of Ki-67 expression in patients with gastric adenocarcinoma. World J Gastroenterol 29:1602–1613

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Funding

This work was supported by Guangzhou Basic and Applied Basic Research Foundation (2023A03J024); National Nature Science Foundation of China (Grant No.82202259); The 13th youth Talent Project of Guangdong Province Hospital of Traditional Chinese Medicine (ZY2022YL05).

Author information

Author notes

  1. Ziqi Jia and Lei Guo share equal contribution as first authors.

Authors and Affiliations

  1. Department of Radiology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, China

    Ziqi Jia, Lei Guo, WenJing Yuan, JianHao Dai, JianYe Lu, ZhiQiang Li, Weicui Chen & Xian Liu

  2. Department of Pathology, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, China

    Xiaohua Du

Authors
  1. Ziqi Jia
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Lei Guo
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. WenJing Yuan
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. JianHao Dai
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. JianYe Lu
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. ZhiQiang Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Xiaohua Du
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Weicui Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Xian Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

ZJ: conceptualization and writing. LG: methodology and data analysis. WY: data curation. JD: visualization. JL: validation. ZL: resources. XD: resources. WC: writing (review and editing). XL: supervision and project administration.

Corresponding authors

Correspondence to Weicui Chen or Xian Liu.

Ethics declarations

Ethics approval and consent to participate

Approval was obtained from the Institutional Review Board of The Second Affiliated Hospital of Guangzhou University of Chinese Medicine.

Consent for publication

Written informed consent was not required for this study because the research presents no more than minimal risk of harm to subjects and involves no procedures for which written consent is normally required outside the research context, such as minimal-risk research that involves surveys/interviews conducted via telephone or online.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Table S1.

MRI sequence parameters of T2WI, DWI, T1WI, and T1-CE of different MR devices. Table S2. Inter-observer reliability for HU tumor, HU intestinal wall and SD muscle. Table S3. Objective image quality analysis.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Jia, Z., Guo, L., Yuan, W. et al. Performance of dual-layer spectrum CT virtual monoenergetic images to assess early rectal adenocarcinoma T-stage: comparison with MR. Insights Imaging 15, 11 (2024). https://doi.org/10.1186/s13244-023-01593-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13244-023-01593-5

Keywords