Leserbrief

With great interest we have read the recent article by T. Rettenbacher describing the use of ultrasound (US) for the assessment of peripheral lymph nodes (Rettenbacher T. Ultraschall in Med 2010; 31: 344 – 362). The article focuses on different aspects of US regarding its ability to better detect, depict, and validate peripheral lymph nodes, especially focuses on the educational aspect and underlines the advantage as compared to clinical examination alone.

This article demonstrates very well the basic appearance of peripheral lymph nodes in US, the variety of possible normal appearances and the diverse patterns that can be useful to detect an involved lymph node. Thus it is an excellent compilation of all details around the lymph node US.

Moreover, the article from Dr. Rettenbacher is full of useful little hints and special features that one can only learn during one’s own learning curve when sitting often enough in front of an US device. We found the description of the normal lymph node in the axilla as centrally echo-free, then echo-rich and again peripherally echo-free very interesting. We call those lymph nodes ”three layer” lymph nodes. Another important description is the observance that there, where the hilum is, the cortex (parenchyma) is withdrawn; a fact that you can always depict but nowhere read about. A further hint is that neck lymph nodes are often without any hilum and that the appearance of their changes over time depending on the age of the patient.

Above all we liked the comparison between the bean model and the lymph node and the explanation which diameter is measured. Just the US pictures should have been exchanged (first part c, then part b, according to the bean model series).

Our own experiences we deduct from, on the one hand, a larger series of skin cancer patients suffering from all kind of skin cancer (melanoma, squamous cell carcinoma, malignant histiocytoma, Merkel cell carcinoma, cutaneous lymphoma, cutaneous involvement of a general lymphoma, Hodgkin disease, etc.) of the skin Cancer Unit of the Charité Berlin and, on the other hand, of the same diversity of diseases presented at the Radiology Department connected with the Skin Cancer Unit in Naples, Italy (Catalano O et al. Curr Probl Diagn Radiol 2010; 39: 30 – 36; Catalano O et al. Voit et al. AJR Am J Roentgenol 2010; 195: 474 – 475). From both our publications we could learn that we described similar, virtually identical patterns without even having known each other before (AJR, 2010, in press). The experience we collect in skin cancer patients cannot automatically be extrapolated to most of the other lymph node involvements. Consequently a lymph node metastasis from different primary tumors may show different spread and growth patterns and consequently different echo-patterns. We may highlight herein how, basing on our experience, melanoma metastases have some peculiarities that should be reminded to the journal reader. First of all we have experience with large prospective series on recurrences of skin cancer patients with corresponding follow-up (Voit C et al. Cancer 2001; 91: 2409 – 2416) and of large prospective studies regarding the estimation of US even in the situation before a sentinel node will be removed (Voit C et al. Ann Surg Oncol 2006; 13: 1682 – 1689; Voit CA et al. J Clin Oncol 2009; 27: 4994 – 5000). From the latter situation we could conclude that the Solbiati index is not helpful for the estimation of sentinel nodes, although we constantly used length to depth ratio (in the article called the ratio of the longitudinal to the short axis diameter, see page 353) meaning that we did the measurements correctly. In this last mentioned prospective study (Voit CA et al. J Clin Oncol 2009; 27:4994 – 5000) even micrometastases in the lymph node could be detected by US. This quantification of metastases could be proven by measuring the diameter of the largest metastases in the sentinel node according to Rotterdam criteria (van Akkooi AC et al. Ann Surg 2008; 248: 949 – 955). Most helpful in our eyes is the combined use of US (as a screening method for detection of suspicious lymph nodes) and US-guided fine needle aspiration cytology (FNAC) (for verification of the previously suspected lesion) in all situations. Thus the often false positive evaluation of a screening modality (US with a lower sensitivity) will be corrected by a rarely false negative method (the FNAC with a very high specificity). Thus you can correct the US false positives into true negatives. Thus a sentinel node metastasis with a nest of 0.5 mm in diameter could be detected by US and even successfully punctured by FNAC (Voit CA et al. J Clin Oncol 2009; 27: 4994 – 5000). This is only possible since the nests measured as that small on the histopathology glass slides, seem to cause alterations in US that are much bigger and can therefore be depicted. We have summarized our observations over the years in a large and ongoing prospective study in which we apply our morphology criteria for detection of sentinel node metastases by US (Voit C et al. J Clin Oncol 2010; 28: 847 – 852), which have recently been published. Here we systematically apply criteria as humps (an irregular broadening of the parenchyma as described by Rettenbacher) or the loss of central echoes (aimed at by Rettenbacher and called there ”the ability to verify the hilum sign” or ”homogeneity of the internal structure”) the balloon shaped lymph node as complete echo-free round terminal stage of lymph node metastases and on the other hand the peripheral perfusion alone or in combination with the B-image criteria as an very early sign.

The peripheral perfusion is also extremely helpful to distinguish fatty tissue inside the medulla from real echo-poor tumor nests. The tumor deposits normally show vascularization due to neo-angiogenesis.

The readers should be aware that modern, high-resolution US scanners and probes allow a very detailed analysis of lymph node metastasis echotexture and angioarchitecture. We encourage them to carefully assess the lymph nodes and even far beyond the classical patterns such as balloon shape and absence of the echogenic hilum. As excellently shown in Dr Rettenbacher’s Fig. 8, US now allows detecting very small tumor deposits within the lymph nodes, if the operator knows that the occurrence of partial metastasis is possible and has the expertise to recognize it (Catalano O et al. J Ultrasound Med 2010; 29: 791 – 802).

Last but not least, seen from the dermatological point of view, we use the typical penniform structure only for description of muscles and describe lipomas as mostly echo-rich with homogenous central echoes, the typical location in the subcutaneous tissue and predominantly showing no perfusion.

The neurinoma is mostly echo-free and without any perfusion and can easily be distinguished from a balloon shaped lymph node with the typical (painful) sensation along the nerve tract during FNAC.

Atheromas do sometimes show their hypoechoic excretory duct, but most of the times present themselves as inflamed structures with even peripheral perfusion and often an onionskin-like B-image.

Thus we just want to summarize that for the overall knowledge about lymph node US the present article by Rettenbacher is excellent, secondly our recently published morphology criteria regarding lymph nodes in malignant melanoma will help to correctly classify the lymph nodes (Voit C et al. J Clin Oncol 2010; 28: 847 – 852) and thirdly that the combination with FNAC (above all with an in house or overnight cytology) will help to verify the lesion. However, we are looking forward to more interesting descriptions in the soon to come up part 2 of the lymph node article.

C.A. Voit, Berlin, Germany
A. C. J. van Akkooi, Rotterdam, Netherlands
O. Catalano, Naples, Italy