British Journal of Radiology 74 (2001),42-47 © 2001 The British Institute of Radiology
Electrical impedance scanning: a new technique in the diagnosis of lymph nodes in which malignancy is suspected on ultrasound
A Malich, MD1,
T Fritsch1,
C Mauch, MD2,
T Boehm, MD1,
M Freesmeyer, MD1,
M Fleck, MD1,
R Anderson3 and
W A Kaiser, MD1
1 Institute of Diagnostic and Interventional Radiology
2 Department of Radiology, Friedrich Schiller University Jena, Bachstrasse 18, D-07740 Jena
3 Siemens Medical Engineering, Special Systems—Women's Health, Henkestrasse 127, D-91052 Erlangen, Germany
 |
Abstract
|
|---|
Differentiation between inflammatory and malignant lymph nodes by ultrasound is difficult. Electrical impedance scanning (EIS) is a new diagnostic tool, so far used primarily for the identification of malignant breast lesions. Cancer cells have altered dielectric properties compared with normal cells, thereby distorting the local electrical field. The induced changes in capacitance and conductivity are measurable using EIS. We evaluated EIS in demonstrating the cause of lymph node enlargement. 51 lymph nodes that were suspicious for malignancy on ultrasound (32 patients, mean age 32 years), with a mean size of 18 mmx12 mmx10 mm, were examined. The following lymph node locations were included in the study: cervical, inframandibular, axillary, paraaortic and inguinal. EIS results were compared with histopathological and follow-up findings. 30/34 malignant lymph nodes were correctly detected using EIS, while 14/17 inflammatory or benign lymph nodes were correctly identified as benign by EIS; thus, there were 4/51 false negative and 3/51 false positive cases. The sensitivity was 88.2% and the specificity was 82.4%. Corresponding negative and positive predictive values were 77.8% and 90.9%, respectively. Results from this initial study suggest the potential usefulness of EIS as an adjunctive imaging modality in the differentiation of lymphadenopathy that is equivocal on ultrasound. The best accuracy was obtained in the cervical, axillary and inguinal regions. Owing to technical restrictions of the present system, examination of inframandibular and paraaortic lymph nodes should be limited to special cases.
|
Introduction
|
|---|
The diagnosis of malignant lymphadenopathy and the differentiation between inflammatory lymph nodes and lymph node relapses is essential in establishing a diagnosis of malignancy, in tumour staging, in therapy planning and also in follow-up during and after therapy [1]. Malignant lymph nodes often show certain typical changes on ultrasound examination such as increased diameter, heterogeneous echogeneity, a particular short-to-long axis ratio, altered appearance of the surrounding soft tissue and posterior enhancement [1, 2]. However, discrimination of small non-necrotic metastases from reactive lymph nodes using ultrasound alone is still very difficult and can produce ambiguous results [1, 3–5]. Vascular changes have been reported to be a more sensitive basis for discriminating between malignant and non-malignant lymph nodes. Even taking this into account, the differentiation methods are still not yet sufficiently accurate.
Electrical differences between benign and malignant tissues have been an active topic of research since the 1920s. Moderate variations in capacitance and conductivity values (inverse of impedance) are reflected by various types of nomal tissue [6, 7]. In contrast to these rather homogeneous observations, malignant tissues demonstrate substantially increased capacitance and conductance, as measured in freshly excised malignant breast tissue [6–8]. These differences are probably attributable to changes in cellular water content, the amount of extracellular fluid, membrane properties and packing density, orientation of malignant cells and possibly the rate of necrosis.
These findings spurred the development of electrical impedance scanning (EIS) as a diagnostic technique. In 1990, Piperno et al [9] published promising initial results of electrical impedance measurement in breast screening. During the last decade, both the technical equipment and the application mode have undergone significant refinement. The currently used EIS-based system, TransScan TS2000 (Siemens, Erlangen, Germany), comprises a real-time, non-invasive and side-effect-free method, primarily intended for classifying mammographically and/or ultrasonically equivocal breast lesions. However, application of this technology is probably not limited to breast diagnosis [10]. Using the new small probe, lymph nodes in various locations are also measurable.
To the best of our knowledge, this is the first report of results from the use of EIS for classification of lymph nodes. Also, to the best of our knowledge, only one application for lesions outside the breast has been described to date. Emtestam et al [11] described the pre-operative assessment of nodular basal cell carcinoma based upon electrical impedance measurements.
The main aim of this study was to evaluate the clinical success of this new technique in classifying lymph nodes in various locations. The following three issues were addressed in the study. (1) To what extent is EIS of value as an additional technique for the classification of suspicious lymph node structures? (2) In which locations could this application be practical? And (3) what is the achievable sensitivity, specificity, false positive rate, false negative rate, positive predictive value and negative predictive value fordetermining the nature of lymph node enlargement?
|
Patients and methods
|
|---|
All patients having a suspicious ultrasound lymph node finding in our institute between April and July 1999 were included in this initial study. A total of 32 patients with 51 ultrasonically equivocal or suspicious swollen lymph nodes in various locations (cervical, axillary, inframandibular, inguinal and paraaortic regions) were included. The ultrasound examinations were performed with a 7.5 MHz transducer on either an HDI 5000 (ATL, USA) or Logic 500 (Kranzbuehler, Germany) system. The size of the lymph nodes was examined by measuring theaxial diameter; a diameter of >5 mm was considered suggestive for pathology. Additionally, the shape and the echogenic hilus were used for qualitative differentiation of the lesion. All lymph node findings underwent a colour and power Doppler mode examination. All lymph nodes showing one or more ultrasound criteria suggestive of malignancy were classified as suspicious. The size and depth of all lesions were measured.
All patients underwent electrical impedance measurement using EIS (TransScan TS2000, TransScan Research & Development Co., Ltd., Israel; distributed by Siemens, Erlangen, Germany), according to normal protocol procedure [12]. (1) A low level (voltage 0.1–2.5 V, alternating current <5 mA, frequency 200 Hz), biocompatible electrical current, applied via a metal cylinder (base electrode) held in the recumbent patient's hand, flows through the patient's body. (2) The hand-held scan probe, containing an 8x8 sensor matrix, is applied to the skin at the location of the suspicious ultrasound finding. Good contact is facilitated by the use of ultrasound gel and ECG spray. (3) The probe sensors measure local electrical current. (4) The computer calculates tissue-related conductivity and capacitance from these quantitative measurements. (5) The display of each parameter is presented separately in real-time as a two-dimensional 256 grey scale image. (6) Up to five recordings are taken at the region(s) of interest (see Figures 1
and 2).
View larger version (66K):
[in this window]
[in a new window]
|
Figure 1. Positive electrical impedance scanning examination of a suspicious cervical lymph node. Spots visible in sectors 1, 2 and 3 represent neoplastic recurrence in the same lymph node; in sector 4, two contralateral neoplastic occurrences are visible.
|
|
View larger version (65K):
[in this window]
[in a new window]
|
Figure 2. Negative electrical impedance scanning of a suspicious cervical lymph node. (Bright lines at the edges of sector 3 represent contact artefacts.)
|
|
An increased value of conductivity/capacitance (decreased impedance) is visible as a bright white focal spot. These "spots", defined as focal brightness in one of the sectors clearly more luminous than its surroundings, were interpreted. A spot shown in conductivity only, or in both conductivity and capacitance, but that does not correspond to a skin lesion, mole or scar and is not determined to be caused by an artefact such as bone, air bubble or contact artefact, was classified as a positive EIS finding and suspicious of malignancy. On average, the examination of one lesion lasted 5–10 min.
Most lesions underwent surgical treatment because of suspicious ultrasound findings, whereas some cases were observed with ultrasound in a short time follow-up of at least 4 weeks. An inflammatory reaction was diagnosed when the lesion was no longer visible after that time, and no other conspicuous clinical observations were made. Similarly, a viral aetiology was assumed if blood serum examination indicated the presence of a viral infection (mononucleosis). In those cases, follow-up examinations by ultrasound were perfomed until the lymph node abnormality was no longer visible. In addition, all available diagnostic techniques (CT, MRI and PET) were used, as necessary, for further discrimination. If, as a consequence of these findings, chemotherapy and/or radiation therapy was prescribed instead of surgical removal, these cases were also scored as malignant. Surgical removal of a malignant lymph node located cervically is not best practice in all cases [13]. Criteria for discrimination between benign and malignant lymph nodes are summarized in Table 1.
The clinical results, as depicted by sensitivity, specificity, false positive rate, false negative rate, positive predictive value and negative predictive value, were calculated with respect to the ultrasonically suspicious lesion.
|
Results
|
|---|
34 of the total of 51 examined lymph nodes were proven malignant and 17 were benign or inflammatory. 38 cases were classified by histopathology reports and 13 cases were classified by follow-up examination. The most common malignant histology was squamous cell carcinoma. The mean size on ultrasound examination and histopathological findings of the lesions are shown in Table 2.
A positive EIS finding (see Figure 1
) was determined in 33 cases, while a negative EIS finding was observed in 18 cases (see Figure 2). 30 lymph nodes were determined as true positive, 4 as false negative, 14 as true negative and 3 as false positive.
The results of EIS are shown in Table 3, subclassified by lymph node location. In this patient population, EIS demonstrated a lymph node examination sensitivity of 88.2% and a specificity of 82.4%. The negative and positive predictive values were 77.8% and 90.9%, respectively (Table 3). The accuracy of EIS results varied depending on the location. The most commonly examined lymph nodes were located cervically (n=25), where a sensitivity of 94.7% (16/17) and a specificity of 67% (4/6) were reached. Of the axillary lesions, 4/5 were true positive and 2/2 true negative (see Table 3). Inguinal lymph nodes were, in this small subset, measured by EIS with perfect accuracy (3/3 true positive and 6/6 true negative). In contrast to these locations, inframandibular and paraaortic lymph nodes yielded false results in 3/8 and 1/2 cases, respectively. Sensitivity and specificity of the particular subsets as well as the total are given in Table 4.
View this table:
[in this window]
[in a new window]
|
Table 3. Correlation between histopathological findings and electrical impedance scanning (EIS) findings of lymph node lesions
|
|
|
Discussion
|
|---|
Although certain morphological differences are seen on ultrasound between malignant lymph nodes and inflammatory lymph nodes, correct differentiation is still difficult [5]. B-mode, colour and power Doppler ultrasound are limited methods and cannot currently replace biopsy and long-term follow-up [14]. We evaluated the potential use of the new technique of EIS in the diagnosis of suspicious lymph nodes. Our experience with lymph nodes shows that EIS achieves a relatively high overall sensitivity of 88.2%, in combination with a reasonable specificity of 82.4%. We consider these values to be promising, particularly in view of the fact that the published sensitivity and specificity values of ultrasound examinations, including colour Doppler ultrasound, range from 66.0–92.8% and 60.0–80.0%, respectively [5, 15–17].
In our initial study, EIS was evaluated as an adjunct to ultrasound. Although colour Doppler mode was used routinely on every lymph node, it was still not possible to unambiguously classify all nodes because of the inconclusive or non-uniform presence of typical sonomorphological criteria. We suggest that in such cases, adjunctive use of EIS is of value in detecting lymph node malignancy. Particularly promising sensitivities (95%) were achieved on cervical lymph nodes.
These initial results, dominated by squamous cell carcinoma, require confirmation by further studies, including larger patient populations and abroader spectrum of pathologies. Areas of interest for future studies include follow-up aftersurgical treatment of cancer, monitoring during chemotherapy and radiation therapy, and examination of lymphomas and melanomas. Additionally, the efficacy of EIS alone is yet to be tested.
In clinical practice, radiologists are commonly faced with deciding between recommending invasive diagnostic procedures (biopsy/surgical removal) or further ultrasound follow-up in the case of indeterminate lymph nodes. In these situations the application of EIS as an adjunctive technique—a patient-friendly, non-invasive and fast real-time examination—should be helpful in patient case management. The current high rate of biopsies of inflammatory lymph nodes, especially in children and adolescents, could thereby be reduced.
Some points need to be taken into account to achieve reliable results: (1) as noted by Malich et al [18], special care must be taken during EIS image interpretation, since air bubbles, bones, muscles, superficial skin lesions, insect bites and poor contact of the scanner can induce high conductivity signals, causing spots similar to those generated by the increased conductivity of malignant cells. These artefacts could be overcome with training and experience. An atlas of normal findings of each lymph node region, with typical bone artefacts, etc., and a direct comparison of every lymph node location in relation to the opposite side would simplify the diagnostic process. Interestingly, benign lymph nodes that grossly protrude the surrounding skin level can cause false positive signals. (2) In general, cervical lymph nodes are more easily accessed, successfully scanned and interpreted than inframandibular lesions on account of their anatomical situation. At present, inframandibular and subparotid locations present the greatest limitations. Interfering bones, contact artefacts and the prominence of some lymph nodes in these locations make the interpretation more difficult and may cause false positive as well as false negative EIS results. This also occurred in some cervical cases and probably caused the aforementioned false positive results. (3) If the lesion is palpable and very superficial, use of EIS is limited owing to contact artefacts caused by the inflexible probe. (4) Applying thescanner on inguinal and axillary lymph nodes gave no contact artefacts. However, the application on inguinal, and sometimes axillary and cervical, lesions can be limited by the volume of surrounding skin hair. The superficial hair should be shaved prior to examination to reduce the potential for false positive findings. (5) The maximum lesion depth at which EIS functions is believed to be approximately 3–3.5 cm, since the local signal distortion caused by the malignancy normalizes as the currents travel. After this distance, the signal measurement technique can no longer distinguish the effect of the distortion. This limitation is not of great importance for the detection of superficial lymph nodes, but appears to limit the correct examination of paraaortic lesions. The two observed paraaortic lymph nodes were located 3 cm and 4 cm deep, as measured by ultrasound. In our small sample size, only one of the two lesions was detectable—allegedly because of its depth. Most suspicious paraaortic findings are at a depth of 6–10 cm, theoretically making them unlikely to be detected by the current EIS system. (6) A more flexible scanner would simplify the examination at almost all locations [18]. Additionally, a scanner with a recording surface smaller than the 65 mm square of the current scanner would allow greatly improved accessibility, which is more important for lymph node examination than for breast scanning.
In conclusion, despite current technical limitations of the breast scanning probe, the EIS system is suitable for use in lymph node examination, initial results being sufficiently promising to warrant larger scale investigations.
Received for publication May 25, 2000.
Revision received August 7, 2000.
Accepted for publication August 31, 2000.
|
References
|
|---|
-
Na DG, Lim HK, Byun HK, Kim HD, Ko YH, Baek JH. Differential diagnosis of cervical lymphadenopathy: usefulness of color Doppler sonography. AJR 1997;168:1311–6.[Abstract/Free Full Text]
-
Ying M, Ahuja AT, Evans R, King W, Metreweli C. Cervical lymphadenopathy: sonographic differentiation between tuberculous nodes and nodal metastases from non-head and neck carcinomas. J Clin Ultrasound 1998;26:383–9.[Medline]
-
Van den Brekel MW, Castelijns JA, Stel HV, Luth WJ, Valk J, van der Waal I, et al. Occult metastatic neck disease: detection with US and US-guided fine needle aspiration cytology. Radiology 1991;180:457–61.[Abstract/Free Full Text]
-
Van den Brekel MW, Castelijns JA, Snow GB. Detection of lymph node metastases in the neck: radiologic criteria. Radiology 1994;192:617–8.[Free Full Text]
-
Ariji Y, Kimura Y, Hayashi N, Onitsuka T, Yonetsu K, Hayashi K, et al. Power Doppler sonography of cervical lymph nodes in patients with head and neck cancer. Am J Neuroradiol 1998;19:303–7.[Abstract]
-
Fricke H, Morse S. The electric capacity of tumors in the breast. J Cancer Res 1926;16:340–76.
-
Jossinet J. The impedivity of freshly excised human breast tissue. Physiol Meas 1998;19:61–75.[Medline]
-
Surowiec A, Stuchly S, Barr R, Swarup A. Dielectric properties of breast carcinoma and the surrounding tissues. IEEE Trans Biomed Eng 1988;35:257–63.[Medline]
-
Piperno G, Frei EH, Moshitzky M. Breast cancer screening by impedance measurements. Front Med Biol Eng 1990;2:111–7.[Medline]
-
Fields SI, Rossman M, Phillips E. Adjunctive improvement of mammographic accuracy using electrical impedance scanning (EIS). Radiology 1998;209:272–3.
-
Emtestam L, Nicander I, Stentström M, Ollmar S. Electrical impedance of nodular basal cell carcinoma: a pilot study. Dermatology 1998;197:313–6.[Medline]
-
TransScan TS 2000 Operation Manual. TransScan Research & Development Co., Ltd., Israel; distributed by Siemens, Erlangen, Germany.
-
Redaelli de Zinis LO, Piccioni LO, Ghizzardi D, Mantini G, Antonelli AR. Indications for elective neck dissection in malignant epithelial parotid tumors. Acta Otorhinolaryngol Ital 1998;18:11–5.[Medline]
-
Adibelli ZH, Unal G, Gul E, Uslu F, Kocak U, Abali Y. Differentiation of benign and malignant cervical lymph nodes: value of B-mode and color Doppler sonography. Eur J Radiol 1998;28:230–4.[Medline]
-
Yusa H, Yoshida H, Ueno E. Ultrasonographic criteria for diagnosis of cervical lymph node metastasis of squamous cell carcinoma in the oral and maxillofacial region. J Oral Maxillofac Surg 1999;57:41–8.[Medline]
-
Giancarlo T, Palmieri A, Giacomarra V, Russolo M. Preoperative evaluation of cervical adenopathies in tumors of the upper aerodigestive tract. Anticancer Res 1998;18:2805–9.[Medline]
-
Giovagnorio F. Doppler color in superficial adenopathies. Radiol Med (Torino) 1999;97:148–52.
-
Malich A, Fritsch T, Freesmeyer MG, Fleck M, Anderson R, Kaiser WA. Electrical impedance scanning (EIS) for classifying suspicious breast lesions—first results. Eur Radiol 2000;10 (2 Suppl. 1):132
This article has been cited by other articles:
|
|
|
|
 
A. Stojadinovic, S. I. Fields, C. D. Shriver, S. Lenington, R. Ginor, G. E. Peoples, H. B. Burch, T. Peretz, H. R. Freund, and A. Nissan
Electrical Impedance Scanning of Thyroid Nodules Before Thyroid Surgery: A Prospective Study
Ann. Surg. Oncol.,
February 1, 2005;
12(2):
152 - 160.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
