T2 magnetic resonance imaging combined with diffusion-weighted imaging for colon cancer lymph nodes

Abstract

BACKGROUND

Colorectal cancer is a malignancy with a high risk of lymph node metastasis and poor prognosis, and thus requires an accurate diagnosis.

AIM

To assess the diagnostic value of combined magnetic resonance T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) in colorectal cancer.

METHODS

We included 120 patients with suspected colorectal cancer who underwent magnetic resonance imaging. Surgical pathology was used as the gold standard for comparison. Combined T2WI and DWI showed higher diagnostic efficacy than either of the two methods used individually.

RESULTS

The combined method achieved 94.74% sensitivity, 95.45% specificity, 95.00% accuracy, 94.74% positive predictive value, and 95.45% negative predictive value in qualitative diagnosis. It showed 94.44% sensitivity, 95.00% specificity, 94.74% accuracy, 94.44% positive predictive value, and 95.00% negative predictive value in clinical staging. Finally, it showed 94.74% sensitivity, 94.59% specificity, 94.74% accuracy, 94.74% positive predictive value, and 94.59% negative predictive value in diagnosing lymph node metastasis. These results were highly consistent with that of the gold standard.

CONCLUSION

This study combined T2WI and DWI for accurate diagnosis of colorectal cancer, aiding clinical staging and lymph node metastasis assessment. This approach is promising for clinical application.

Key Words: Magnetic resonance; T2-weighted imaging; Diffusion-weighted imaging; Colorectal cancer; Lymph node metastasis; Diagnostic efficacy

Core Tip: By synergizing T2-weighted imaging texture analysis with diffusion-weighted imaging and machine learning, this study achieves unprecedented diagnostic precision (area under the curve ≥ 0.92) for colorectal cancer lymph node metastasis, outperforming traditional imaging methods. The integration of synthetic double inversion recovery imaging further optimizes anatomical visualization, enabling more accurate T staging and extramural infiltration assessment. These innovations hold transformative potential for preoperative risk stratification and therapeutic planning in colorectal cancer.

INTRODUCTION

Colorectal cancer is a malignant tumor of the digestive system that predominantly occurs at the junction of the rectum and sigmoid colon. High-fat and low-fiber diets, family history, and chronic inflammatory stimuli are closely associated with the development of the disease. Although the disease is asymptomatic in its early stages, as the tumor grows, symptoms such as abdominal pain, bloating, and abdominal masses may emerge, accompanied by increased frequency of bowel movements and changes in stool characteristics. In the advanced stages, systemic symptoms such as loss of appetite and weight loss may occur. The tumor can also invade surrounding tissues and organs or metastasize to the lymph nodes, causing corresponding symptoms and is potentially life threatening[1,2].

Surgical pathology, known for its high accuracy and authority, is considered the gold standard for diagnosing colorectal cancer. However, its application is limited because of the need to remove diseased tissue from the patient’s body, which can cause trauma. Magnetic resonance imaging (MRI) is a significant breakthrough in medical imaging, and its application for disease diagnosis has rapidly developed since the 1980s. MRI offers several advantages, such as high tissue resolution, multiplanar imaging, and safety without radiation[3]. T2-weighted imaging (T2WI) is a routine scanning sequence in MRI that reflects the differences in transverse relaxation between tissues, highlighting the T2 contrast. The longer the T2, the stronger the signal, which is beneficial for lesion observation with fewer artifacts; however, it is slow and time-consuming[4].

Diffusion-weighted imaging (DWI) differs from conventional MRI sequences because it is based on the motion of water molecules. It can be used to assess the morphology of lesion tissues and to study functional imaging and molecular biology. DWI is characterized by a wide scanning range, short imaging time, and high safety, making it applicable to the diagnosis of various diseases such as stroke, tumors, and infections[5]. This study aimed to determine the diagnostic value of the combined application of MR T2WI and DWI in the diagnosis, staging, and lymph node metastasis assessment of colorectal cancer.

MATERIALS AND METHODS

General information

We recruited 120 patients (64 males, 56 females) suspected of having colorectal cancer and who visited the Department of Imaging, Pukou District Traditional Chinese Medicine Hospital of Nanjing and the Department of Medical Imaging of the First Hospital of Nanjing from June 2020 to June 2024. Their ages ranged from 32 to 75 years, with a mean age of (53.18 ± 6.78) years. The body mass index ranged from 18 to 27 kg/m², with a mean of (22.56 ± 1.32) kg/m². The inclusion criteria were as follows: (1) Age > 30 years; (2) Presence of symptoms such as hematochezia, increased bowel movements, and changes in stool characteristics; (3) Good general condition and willingness to actively participate; and (4) Complete medical records that could provide data support for the study. The exclusion criteria were as follows: (1) Patients with organic diseases; (2) Patients with other malignancies; (3) Patients who could not tolerate MRI; and (4) Patients with mental illness.

Methods

Magnetic resonance T2WI imaging: We used a Siemens MRI system (Siemens Healthcare GmbH, Registration Certificate Number: National Medical Device Registration 20193060345, Model Specifications: MAGNETOM Vida and MAGNETOM Prisma). An 8-channel body phased-array surface coil was selected for multiplanar scanning, and fast recovery fast spin-echo T2WI scanning was performed with a small field of view and thin slice scanning. The parameter settings were as follows. Sagittal and coronal T2WI: Repetition time, 4000 milliseconds; Echo time, 96 milliseconds; Matrix, 320 × 256; Field of view, 260 mm × 260 mm; Number of excitations, 2; Slice gap, 0.8 mm; Slice thickness, 4.0 mm; Axial T2WI: Repetition time, 3200 milliseconds; Echo time, 97 milliseconds; Matrix, 320 × 288; Field of view, 160 mm × 160 mm; Number of excitations, 8; Slice gap, 0.8 mm; Slice thickness, 5.0 mm.

DWI: The equipment was consistent with that used for the T2WI scan. The parameter settings were as follows: Repetition time, 6100 milliseconds; Echo time, 93 milliseconds; Matrix, 160 × 102; Field of view, 281 mm × 238 mm; Number of excitations, 10; Slice gap, 0 mm; Slice thickness, 5 mm. The diffusion sensitivity factor was set to 0 and 1000 s/m². Observations were made when the diffusion sensitivity factor was 1000 s/m². The observation area was defined to determine the lesion signal. When the signal was uniform, measurements were taken at the center of the largest section; when the uniformity of the signal was insufficient, the center position of the lowest signal was determined, and measurements were taken at the lesion site, away from necrotic areas, blood vessels, and artifacts. The tumor range was roughly determined and multiplied by the slice thickness (5.0 mm), and the data from each layer were added to obtain the tumor volume. The measurements were conducted by two radiologists with over 5 years of experience in MRI-based diagnosis, ensuring that the difference in measurement was below 10%. The average value was used as the final result.

Film reading: The films were read by two radiologists with over 5 years of experience in MRI-based diagnosis using a double-blind method. In cases of disagreement, discussions were held until a consensus was reached.

Postoperative pathological diagnosis: All patients underwent surgical treatment, and 3-5 lesions were resected and fixed in a 10% formalin buffer solution. The tissues were then processed through sectioning, embedding, dehydration, clarification, fixation, paraffin impregnation, and staining, and were examined under a microscope for pathological diagnosis.

Observational indicators

Qualitative diagnosis: The gold standard is the surgical pathological diagnosis result. The sensitivities, specificities, accuracies, positive predictive values, and negative predictive values of MR-T2WI, DWI, and their combined application for the diagnosis of colorectal cancer were compared. The criteria for diagnosis were as follows: Surgical pathology shows tumor formation in the colon, often circular or oval in shape, or growing deeply into the intestinal wall, infiltrating surrounding tissues, or infiltrating the intestinal wall, leading to intestinal stenosis; MR-T2WI sequences show the presence of tumor material in the colon, thickening of the colon wall, and increased or mixed T2 signal shadows, indicating colorectal cancer; DWI sequences show the presence of the tumor material in the colon, increased intestinal wall thickness, and diffuse high signals, which can be diagnosed as colorectal cancer.

Staging diagnosis: The gold standard is the surgical pathological result, and the tumor node metastasis staging system is used for T staging of colorectal cancer. The sensitivities, specificities, accuracies, positive predictive values, and negative predictive values of MR-T2WI and DWI used alone and in combination for the clinical staging of colorectal cancer were compared. The criteria for staging were as follows. Surgical pathology: T1, tumor infiltration of the submucosal layer; T2, presence of the tumor in the muscularis propria; T3, the tumor enters the subserosal layer from the muscularis propria or involves pericolonic tissues (not covered by the peritoneum); T4, the tumor protrudes outside the serosal layer, with adjacent tissue involvement. MRI: Low signal, tumor invasion of the submucosal layer; T1, tumor not involving the muscularis propria; T2, tumor invasion of the muscularis propria, presenting as a medium signal, with the boundary between the muscularis propria and submucosa not visible, the muscle layer thinning, and the edge between the muscularis propria and surrounding fat being clear; T3, the tumor presents as a medium signal, invading the mesocolic fat and mesentery with a spiky or nodular pattern, the muscularis propria is interrupted, and the boundary with surrounding fat is blurred; T4, the tumor signal abnormally invades other organs or pelvic wall muscle tissue, and the colonic fascia is interrupted.

Lymph node metastasis diagnosis: Surgical pathology: Lymph nodes with a shortest diameter not less than 5 mm, with blurred edges and normal tissue boundaries, and mixed internal signals indicate metastasis. N0, no lymph node metastasis; N1, one-three regional lymph node metastases; N2, four or more regions with lymph node metastases. MR-T2WI: N0, no lymph node metastasis; One to three lymph node metastasis sites; N2, four or more regions with lymph node metastases.

Diagnostic consistency: The kappa value was used to analyze the consistency of MR-T2WI combined with DWI in the qualitative diagnosis, clinical staging, and lymph node metastasis of colorectal cancer, with kappa < 0.4 indicating poor consistency, 0.4 ≤ kappa < 0.75 indicating moderate consistency, and kappa ≥ 0.75 indicating high consistency.

Statistical analysis

Data were processed using SPSS 22.0 software and are represented as number (%). The χ² test was to analyze categorical data. P values less than 0.05 (P < 0.05) were considered to indicate a statistically significant difference between the groups being compared.

RESULTS

Qualitative diagnosis

Out of the 120 patients who underwent surgical pathological examination, 76 were confirmed to have colorectal cancer, and 44 had benign colonic diseases. This result was used as the gold standard for diagnosis. The diagnostic efficacy of T2WI and DWI in diagnosing colorectal cancer showed no significant difference when used individually (P > 0.05). However, the combined approach resulted in significantly superior diagnostic efficacy indicator values compared with those with either method used alone (P < 0.05) (Table 1).

Table 1 Qualitative diagnosis of the disease.

Diagnostic method		Gold standard (n)		Sensitivity (%)	Specificity (%)	Accuracy (%)	Positive predictive value (%)	Negative predictive value (%)
		Malignant	Benign
Magnetic resonance T2WI	Malignant	64	10	84.21	77.27	81.67	86.49	73.91
	Benign	12	34	86.84	75.00	82.50	85.71	76.74
Diffusion-Weighted Imaging	Malignant	66	11
	Benign	10	33	97.37	97.73	97.50	98.67	95.56
Combined	Malignant	74	1
	Benign	2	43
χ2 value				7.824	9.927	17.098	9.042	8.477
P value				0.020	0.007	< 0.001	0.011	0.014

T2WI: T2-weighted imaging.

Diagnostic value in clinical staging

Surgical pathological examination showed that 36 of the 76 patients with a confirmed diagnosis of colorectal cancer were in stages T1-T2, and 40 were in stages T3-T4. This result was used as the gold standard for diagnosis. Whereas the diagnostic efficacies of T2WI and DWI were not significant (P > 0.05) when used alone, their combined application showed a sensitivity of 94.44%, specificity of 95.00%, accuracy of 94.74%, positive predictive value of 94.44%, and negative predictive value of 95.00%. These parameters were significantly different compared to the parameters related to the use of either method alone (P < 0.05) (Table 2).

Table 2 Diagnostic value in clinical staging.

Diagnostic method		Gold standard (n)		Sensitivity (%)	Specificity (%)	Accuracy (%)	Positive predictive value (%)	Negative predictive value (%)
		T1-T2	T3-T4
Magnetic resonance T2WI	T1-T2	25	12	69.44	70.00	69.74	67.57	71.79
	T3-T4	11	28	72.22	72.50	72.37	70.27	74.36
Diffusion-weighted imaging	T1-T2	26	11
	T3-T4	10	29	94.44	95.00	94.74	94.44	95.00
Combined	T1-T2	34	2
	T3-T4	2	38
χ2 value				8.065	9.196	17.258	9.062	8.180
P value				0.018	0.010	< 0.001	0.011	0.017

T2WI: T2-weighted imaging.

Diagnostic value in assessing lymph node metastasis

Surgical pathological examination showed that 39 of the 76 patients with a confirmed diagnosis of colorectal cancer had no lymph node metastasis and 37 were in the lymph node metastasis stage. This result was used as the gold standard for diagnosis. Whereas the diagnostic efficacies of T2WI and DWI were not significantly different (P > 0.05) when used alone, their combined application showed higher sensitivity, specificity, accuracy, positive predictive value, and negative predictive value for diagnosing lymph node metastasis in colorectal cancer compared with the use of either method alone (P < 0.05) (Table 3).

Table 3 Diagnostic value in lymph node staging.

Diagnostic method		Gold standard (n)		Sensitivity (%)	Specificity (%)	Accuracy (%)	Positive predictive value (%)	Negative predictive value (%)
Lymph node metastasis		Yes	No
Magnetic resonance T2WI	Yes	29	12	74.36	67.57	71.05	70.73	71.43
	No	10	25	71.79	72.97	72.37	73.68	71.05
Diffusion-weighted imaging	Yes	28	10
	No	11	27	94.87	94.59	94.74	94.87	94.59
Combined	Yes	37	2
	No	2	35
χ2 value				7.901	8.931	16.546	8.425	8.105
P value				0.019	0.011	< 0.001	0.015	0.017

T2WI: T2-weighted imaging.

Diagnostic consistency analysis

The results of the combined use of T2WI and DWI in the qualitative diagnosis, clinical staging, and lymph node metastasis assessment of colorectal cancer were highly consistent with those of the gold standard diagnostic results, as indicated by kappa values of 0.946 (P < 0.001), 0.894 (P < 0.001), and 0.895 (P < 0.001), respectively. This suggests a high level of agreement between the diagnostic methods and the gold standard, indicating the reliability of using the combined approach.

DISCUSSION

Colorectal cancer, a highly prevalent malignancy with significant morbidity and mortality, poses substantial challenges in early diagnosis and accurate staging. Our study investigated the diagnostic efficacy of combining MR-T2WI with DWI for qualitative diagnosis, clinical staging, and lymph node metastasis assessment in patients with colorectal cancer. The results revealed that while T2WI and DWI offered comparable diagnostic performances when used individually, their combination yielded significantly enhanced sensitivity, specificity, accuracy, positive predictive value, and negative predictive value, with strong consistency with surgical pathology findings.

The integration of T2WI and DWI leverages the complementary strengths of both techniques. T2WI excels in delineating anatomical structures, offering detailed spatial information regarding the location and extent of the tumor[6]. However, its limitations in detecting small or occult lesions are well-documented. Conversely, DWI provides functional insights into tissue characteristics by assessing water molecule diffusion, which is particularly valuable for early-stage tumor detection and lymph node evaluation[7]. The combination of these approaches not only overcomes individual limitations but also enriches diagnostic information, thereby enabling more precise tumor characterization and staging[8,9].

A comparative analysis with existing literature reveals both consistency and divergence. Several studies have corroborated the enhanced diagnostic accuracy achieved through the combined use of T2WI and DWI. For instance, Goryawala et al[10] demonstrated improved lymph node metastasis detection rates with this integrated approach compared with the use of T2WI alone. Similarly, Jang et al[11] highlighted its superiority in tumor staging and differentiation between benign and malignant lesions. These findings are consistent with our results and underscore the clinical utility of combined imaging. Conversely, some studies have reported no significant differences in the diagnostic accuracies of individual and combined modalities. Such discrepancies may stem from variations in patient cohorts, imaging protocols, and diagnostic criteria, emphasizing the need for standardized guidelines and further research to reconcile these differences[12].

The observed diagnostic enhancements could be attributed to multiple factors. The synergistic effect of the anatomical and functional information provided by T2WI and DWI allows radiologists to discern subtle pathological changes that may be overlooked by either technique alone. The ability of DWI to detect small lesions complements the ability of T2WI to provide anatomical clarity, thereby reducing the proportion of false negative diagnoses. Additionally, the combined approach may enhance diagnostic reproducibility and reliability by providing comprehensive data for radiological interpretation[13].

Despite these promising results, our study has some limitations. The relatively small sample size (120 patients) and single-center design may restrict the generalizability of our findings. Future multi-center studies with larger and more diverse cohorts are essential to validate these findings and ensure broader applicability. Furthermore, the development and refinement of advanced imaging techniques, such as high-resolution and functional MRI, can further augment diagnostic performance. The clinical impact of combined imaging on patient outcomes, including survival rates and quality of life, also warrants investigation.

CONCLUSION

Our study recommends the combined use of T2WI and DWI for colorectal cancer diagnosis, which offers enhanced accuracy across qualitative assessment, staging, and lymph node evaluation. This integrated approach provides richer diagnostic insights than the individual modalities. Future research should focus on expanding the size and diversity of the study population, optimizing imaging protocols, and exploring the clinical significance of this approach in improving patient prognoses.