Imaging features of sentinel lymph node mapped by multidetector-row computed tomography lymphography in predicting axillary lymph node metastasis

Xiaochan Ou, Jianbin Zhu, Yaoming Qu, Chengmei Wang, Baiye Wang, Xirui Xu, Yanyu Wang, Haitao Wen, Andong Ma, Xinzi Liu, Xia Zou, Zhibo WenBMC Medical Imaging 2021 · DOI

TL;DR

This study explores the use of multidetector-row computed tomography lymphography (MDCT-LG) to predict axillary lymph node (ALN) metastasis in breast cancer patients. By analyzing specific imaging features of sentinel lymph nodes (SLNs), particularly cortical thickness, the researchers found a reliable method to identify metastatic nodes, potentially reducing the need for invasive sentinel lymph node biopsies in some cases.

What Problem Does This Paper Address?

Accurately determining whether breast cancer has spread to the axillary lymph nodes (ALNs) is crucial for treatment planning and patient prognosis. Traditionally, sentinel lymph node biopsy (SLNB) has been the gold standard for staging, but it's an invasive procedure with potential complications like lymphedema. While imaging techniques like ultrasound and MRI are used, they have limitations in accurately assessing ALN status, sometimes leading to overtreatment or undertreatment. This paper addresses the need for a non-invasive, accurate method to predict ALN metastasis, specifically focusing on the sentinel lymph node (SLN) identified through a specialized CT imaging technique.

Methods

The study employed multidetector-row computed tomography lymphography (MDCT-LG) to visualize SLNs. After injecting a contrast agent into the breast tissue, a high-resolution CT scan was performed. The researchers then used 3D reconstruction software to identify and analyze SLNs. Key imaging parameters measured included:

  • Long-axis diameter: The longest dimension of the SLN.
  • Short-axis diameter: The shortest dimension of the SLN.
  • Ratio of long-/short-axis: A measure of the SLN's shape.
  • Cortical thickness: The thickness of the outer layer of the SLN.

These measurements were then correlated with the pathological findings of ALN metastasis. Logistic regression and Receiver Operating Characteristic (ROC) curve analysis were used to determine the predictive value of these imaging features for ALN metastasis.

Datasets

The study included 112 female breast cancer patients who underwent preoperative MDCT-LG. These patients were diagnosed between January 2019 and March 2021. The dataset was divided into metastatic and non-metastatic ALN groups based on subsequent surgical pathology.

Key Insights and Ideas

  • Cortical Thickness as a Key Predictor: The most significant finding is that the cortical thickness of the SLN, as visualized by MDCT-LG, is a strong independent predictor of ALN metastasis. This suggests that changes in the SLN's outer layer are a direct indicator of cancer spread.
  • Potential to Avoid SLNB: The study's findings suggest that MDCT-LG, particularly the measurement of cortical thickness, could help identify patients with a low probability of ALN metastasis. This opens the door to potentially omitting SLNB in select cases, thereby reducing the invasiveness of treatment.
  • SLN Location is Predictable: The study confirms that SLNs are typically located in Level I of the axilla, aligning with previous research on lymphatic drainage patterns in breast cancer.
  • Combined Features Don't Necessarily Improve Accuracy: While multiple imaging features were analyzed, combining them did not significantly improve the diagnostic accuracy (AUC) compared to using cortical thickness alone. This simplifies the diagnostic approach.

Key Results

  • Cortical Thickness Difference: Metastatic SLNs had a significantly thicker cortex (4.0±1.2 mm) compared to non-metastatic SLNs (2.4±0.7 mm) (P<0.001).
  • Independent Predictor: Multivariate analysis identified cortical thickness > 3.3 mm as an independent predictor of ALN metastasis, with an Odds Ratio (OR) of 24.53 (95% CI 6.58–91.48, P < 0.001).
  • Optimal Cut-off for Cortical Thickness: A cortical thickness of > 3.3 mm yielded the best performance for predicting ALN metastasis, with a sensitivity of 76.2%, specificity of 88.5%, and an Area Under the Curve (AUC) of 0.872 (95% CI 0.773–0.939, P<0.001).
  • Other Features: Long-axis diameter and short-axis diameter were also significantly associated with ALN metastasis, but cortical thickness emerged as the strongest independent predictor.

Limitations and Future Work

The authors acknowledge several limitations:

  • Paired Comparison Difficulty: Potential biases exist due to the difficulty in directly comparing MDCT imaging findings with pathological results for each individual lymph node.
  • Retrospective and Small Sample Size: The study is retrospective and conducted at a single institution with a relatively small sample size, which may limit the generalizability of the findings.
  • Image Quality and Interpretation: The accuracy of measurements can be affected by image quality and the interpretation of fuzzy preprocessing techniques might be necessary.

Future work should focus on prospective clinical trials to validate these findings. Developing improved scanning methods and parameters for ALN assessment is also suggested. The authors hope that future research will lead to more accurate selection of patients for direct axillary lymph node dissection (ALND) or those who can be exempted from SLNB.

Summary updated

BibTeX
@article{ou2021imaging,
  author = {Xiaochan Ou and Jianbin Zhu and Yaoming Qu and Chengmei Wang and Baiye Wang and Xirui Xu and Yanyu Wang and Haitao Wen and Andong Ma and Xinzi Liu and Xia Zou and Zhibo Wen},
  title = {{Imaging features of sentinel lymph node mapped by multidetector-row computed tomography lymphography in predicting axillary lymph node metastasis}},
  journal = {BMC Medical Imaging},
  year = {2021},
  doi = {10.1186/s12880-021-00722-0},
}
Figure