Color Doppler ultrasonography is the only accurate method of displaying the angioarchitecture of lymph nodes. A distorted angioarchitecture has been described as occurring more frequently in malignant lymph nodes. Tschammler and associates evaluated reactive and malignant lymphadenopathy using Doppler ultrasonographic studies of intranodal blood vessels.

Doppler ultrasonography of 117 lymph nodes was performed in 100 consecutive patients with lymphadenopathy before surgical biopsy, neck dissection or high-speed core biopsy. The intranodal architecture was assessed by color Doppler. At least one lymph node from each patient was studied. Most lymph nodes were located in the cervical, submandibular or supraclavicular regions. After excision, the size and shape of the nodes were compared with the ultrasound results to ensure that exactly the same nodes were assessed by both ultrasonography and histology. Specific criteria for the angioarchitecture were established to allow precise differentiation of benign and malignant findings.

Histologic examination yielded 48 reactive lymph nodes, 56 lymph node metastases and 13 malignant lymphomas. Malignant lymph nodes had larger diameters in all axes than did reactive lymph nodes. A missing echogenic center and a round shape were further signs of malignancy. Of the 69 malignant lymph nodes, 96 percent met at least one criterion of malignancy in the assessment of the intra-nodal angioarchitecture. Of the 48 reactive lymph nodes, 25 percent showed no perfusion, 52 percent had normal angioarchitectural features, and 23 percent were classified as false-positives.

Lymph nodes that were classified as malignant on the basis of color Doppler ultrasonography had an average of 2.6 criteria of malignancy. The negative predictive value was 93 percent, and the positive predictive value was 86 percent when at least one criterion of malignancy was present. If all four criteria of malignancy were present, the positive predictive value increased to 94 percent. The echogenicity of the center of the lymph nodes representing interfaces between tissue and vessels was the only sonomorphologic criterion that was correlated with the angioarchitecture.

Results of this study showed that if intra-nodal vessels are assessed carefully, reactive lymph nodes display the same angioarchitecture known from histopathologic studies. The false-positive results were most likely due to overinterpretation of peripheral flow signals depicted on the high-resolution equipment. One of the three false-negative results showed a low-grade non-Hodgkin's lymphoma. The two others occurred in cases of metastases of squamous cell carcinoma. It is suggested that the false negativity was caused by regressive tissue alterations that yielded low perfusion.

The authors conclude that the use of color Doppler ultrasonography in the assessment of intranodal angioarchitecture in superficial lymph nodes is a reliable and reproducible method of differentiating between reactive and malignant lymphadenopathy. With color Doppler ultrasonography, a total of 88 percent of nodes were classified correctly, resulting in a specificity of 77 percent and a sensitivity of 96 percent.