Int. Med J Vol. 6 No 2 December 2007

A New Diagnostic Approach in the Detection of Breast Carcinoma: Electrical Impedance Measurement with TransScan TS 2000

Amran AR¹, M. Fatimah²

¹ Department of Radiology, Kulliyyah of Medicine, International Islamic University of Malaysia, Kuantan, Malaysia.

 ²  Department of Radiology, University of Malaya Medical Centre, Kuala Lumpur, Malaysia.

ABSTRACT

Introduction: Mammography is commonly regarded as the single most important tool for screening and for early detection of breast cancer. However it is not generally recommended for women under 40 years of age and in those taking hormone replacement therapy as the increased density of the breast parenchyma may make mammography more difficult to read and interpret. The limitations of mammography have spurred attempts to find new techniques that can be used either separately or in conjunction with mammography. Purpose: The aim of this study was to quantify the clinical value of using electrical impedance scanning (EIS) or Trans Scan as an adjunct to mammography in order to identify cancerous tissue based upon its inherent altered local dielectric properties. Methods and Materials: The patients were examined using Trans Scan (Trans Scan Medical, Ltd., distributed by Siemens AG. The study population was derived from patients with suspicious breast lesions categorized as BIRADS 3 or 4 detected during mammography or ultrasound. Results: Fifty-three women with 53 mammographically and/or sonographically suspicious findings were examined using EIS. With respect to the histopathological findings (15 malignant and 38 benign lesions) 13 of 15 (86.6% sensitivity) malignant lesions were correctly identified using EIS whereas, 33 of 38 (81.5% specificity) benign lesions were correctly identified. Negative and positive predictive values of 93.9% and 65% were observed respectively. Two benign lesions were correctly identified in a dense breast. The smallest lesion detected in this study measured 20 x 14 mm, which was an infiltrating ductal carcinoma. Conclusion: Electrical impedance scanning as an adjunct to mammography or ultrasound in classifying suspicious lesions is promising because it increases the sensitivity for cancer detection and may reduce biopsy of equivocal lesions.  The additional use of EIS with negative predictive value of 93.9% may be useful to exclude some benign lesions from further diagnostic or invasive procedures. Artifacts, such as signals from superficial skin lesions, poor contact and bubbles are currently a limitation.  

KEYWORDS: breast cancer, mammography, electrical impedance scanning.

INTRODUCTION

Breast cancer is the most common cancer in women in Malaysia and has high morbidity and mortality. Statistically, one in nineteen women in Malaysia is affected by carcinoma of the breast during her lifetime with a currently rising incidence [1]. Despite the low risk of breast cancer in women under the age of 40, a rising trend is noted in this age group. Breast carcinomas in young women tend to be more aggressive than those in older women. Therefore, early detection of breast cancer in this age group is important. Currently the only used screening tool for young women is clinical breast examination (CBE). However, CBE can only detect a tumour that have grown large enough to discern by touch and at which point they are more difficult to treat effectively.

Mammography is commonly regarded as the single most important tool for screening and for early detection of breast cancer. Mammography plays   a major role in the early detection of breast carcinoma. An estimated that a 30% reduction in mortality is noted in association with screening of women aged 40-74 years [2]. However it is not generally recommended for women under 40 years of age and in those taking hormone replacement therapy as the increased density of the breast parenchyma may make mammography more difficult to read and interpret. The limitations of mammography have spurred attempts to find new techniques that can be to be used either separately or in conjunction with mammography.

Breast ultrasound is frequently used to evaluate breast abnormalities that are found on screening or diagnostic mammography or during a clinical breast examination by a physician. Ultrasound is generally accepted as the method of choice for differentiation of cysts from solid masses and guidance in interventional procedures. Ultrasound is also very operator-dependent. In addition, ultrasound cannot document how much breast tissue has been imaged; therefore it is difficult to evaluate the thoroughness of the exam. Ultrasound is most valuable when used after mammography or physical breast examinations have indicated an abnormality.  

Electrical impedance scanning is the first completely new imaging method to receive USFDA clearance since magnetic resonance imaging was approved in 1984 [3]. Electrical impedance is a measurement of how electricity travels through a given material. Every tissue has different electric impedance determined by its molecular composition. Some materials have high electrical impedance while others have low electrical impedance. Recent in vitro measurements of electrical properties of breast tissue show that malignant tissue has much lower electrical impedance or higher electrical conductivity than normal tissue or benign tumours [4]. These dramatic shifts in electric impedance results from cytological and histological changes involving cellular water content, amount of extracellular fluid, membrane properties, packing density and orientation of cells that are typical of malignant tissue transformation [5]. Electrical impedance scanning measures low-level bioelectric current to produce real time image with electrical impedance properties. The resulting impedance images of breast tissue can be used to determine if the region of interest is normal or has a cancerous tumour. Supplementing mammography with EIS or other imaging examinations (such as ultrasound) may reduce the number of unnecessary biopsies and create savings in healthcare costs 

OBJECTIVES

General Objective

The primary goal of this study is to evaluate the clinical accuracy of this new imaging modality in the detection of breast tumour.

Specific Objectives

1) To evaluate if electrical impedance can differentiate between benign and malignant breast lesions

2) To determine false positive findings in normal control group with electrical impedance scanning

METHODS

Clinical Data

This prospective study was conducted at the University Malaya Medical Center between November 2001 and December 2003. The patients were examined using Trans Scan (Trans Scan Medical, Ltd., distributed by Siemens AG. The author and his supervisor reviewed the results.

The study population was derived from patients with suspicious breast lesions categorized as BIRADS 3 or 4 detected during mammography or ultrasound. A total of 53 patients underwent electrical impedance scanning or Trans Scan examination and the findings were compared with biopsy results. Of the 53 patients, there were 15 Malays (28%), 24 Chinese (46%) and 14 Indians (26%).

A second group of patients comprising 69 patients and who were normal clinically, on ultrasound or at mammography acted as a control group. This was also used to look for false positive results. These patients were subjected to four quadrant Trans Scan examination of both breasts. Of the 69 patients, there were 33 Malays (48%), 20 Chinese (29%) and 16 Indians (23%).

Technique of electrical impedance scanning

The patients were informed about the planned EIS examination in detail. After clarification, all patients agreed to the additional examination and underwent an EIS examination according to the normal protocol procedure. A low-level (voltage 0.1-2.5 V, alternating current < 5mA, frequency below 3.5 kHz), biocompatible electrical current was applied via a metal cylinder held in the recumbent patient’s hand, and flowed through the patient’s body. A hand-held scanning probe, comprising of an 8 x 8 mm sensor matrix, was applied to the skin at the location of the suspicious mammography finding. Good contact was facilitated with the used of ultrasound gel. The probe sensors measured local electric current over a band of frequencies. The computer used these quantitative measurements to calculate tissue-related conductance and capacitance. The display of each parameter was separately presented in real-time as a two-dimensional 256 gray scale image. Up to four recordings were taken at the regions of interest.

Factors taken into consideration for the study were:

1.   Category of breast lesion based on the site, shape, size and number

2.   Age of the patient

3.   Race of the patient

Exclusion criteria

1.   Patients who had previous breast biopsy

2.  Patients who has pacemakers

3.  Pregnant patients

Data Analysis

We regarded the histopathological result as the gold standard, to which our Trans Scan results were compared. The resulting sensitivity, specificity, false positive rate, false negative rate, positive predictive value, negative predictive value and accuracy were evaluated with respect to the suspicious lesion.

The author and his supervisor reviewed the Trans Scan images.

Measurements were taken at different angles to the lesion for patients with suspicious breast lesions and four quadrants measurements for patients with normal mammography or ultrasound findings. 

Statistical Analysis

Testing of statistical significance was performed using paired samples t-test, (SPSS). A p-value of <0.05 was considered significant.

 RESULTS

As at the end of November 2003 a total of 53 cases with suspicious lesions detected on

mammography and ultrasound had Trans Scan examination followed by biopsy and 69 cases were used as the  control group.

1. Results of patients with suspicious lesions who underwent Trans Scan examination

Of the 53 patients 15 (28%) were Malays, 24 (46%) were Chinese and 14 (26%) were Indians. Age ranged from 19 to 78 years with the mean age being 45.5 years.

Twenty lesions were positive on EIS as interpreted by a focal increased of conductivity, visible as bright spot. The remaining thirty-three examinations did not show a focal increase, leading to a negative finding.

The histopathological findings, number of cases and mean sizes of the lesions on mammography and ultrasound were observed as shown in Table 1. Histopathological findings of all lesions with false positive EIS is shown in Table II and lesions with false negative EIS are shown in Table III.

3. Analysis of conductance and capacitance values in the study population

There is significant difference (p<0.05) of level of suspicious between conductance and capacitance values of both benign and malignant breast lesions detected using electrical impedance scanning as shown. Boxplots of level of suspicious between conductance and capacitance values of both benign and malignant breast lesions were shown in Figure I and II. This significant difference is of key importance for TransScan to differentiate benign from malignant breast lesions in this study. Figure III showed the electrical impedance scanning images in high-resolution targeted mode, from a patient with histological diagnosis of fibroadenoma. No abnormal signals were detected with low level of suspicion of malignancy. Figure IV showed electrical impedance scanning images in high-resolution targeted mode, from a patient with histological diagnosis of intraductal carcinoma. The abnormal signals were detected with high level of suspicion of malignancy.

2. Results of patients with normal mammography or ultrasound examination who acted as a control group

There were 69 patients, of which 33 (48%) were Malays, 20 (29%) are Chinese and 16 (23%) were Indians. Age of patients ranged from 19 to 71 years with a mean age of 43 years. 48 patients showed normal electrical impedance values whereas 21 patients showed increased impedance values that led to false positive results. Figure V showed the electrical impedance scanning images taken from patients who acted as a control group.

DISCUSSION

The United States Food and Drug Administration (USFDA) approved EIS in 1999 to be used as an adjunctive tool to mammography. From the time of its description, nearly 20 papers have been published on EIS. All studies show promising results in providing confident diagnosis and reducing the number of unnecessary biopsies. EIS can also confirm if the lesion is benign [6-15].

Our study revealed that EIS showed high sensitivity and specificity in detecting lesions deemed equivocal at mammography or an ultrasound as showed in Table IV. Comparison of recent studies done by various investigators and our study are as shown in Table V. Our study on EIS showed the results are comparable to previous studies. Compared to mammography (with regards to histopathological results as a gold standard) EIS demonstrated 86.6% sensitivity and 81.5% specificity. The control group showed specificity of 70%.

In this study two benign lesions were correctly identified (benign proliferative disease and fibroadenoma) in a dense breast. Dr Scott Fields, one of the researchers involved in the clinical trial of EIS found that impedance scanning works well for small lesions in younger patients with dense breasts [7]. This is important in the detection of breast carcinoma in younger patients or patients on HRT in which mammography is not sensitive. In younger patients, the combination of EIS and ultrasound will increase the detection, sensitivity and specificity for malignant breast lesions without resorting to mammograms.

EIS is also better tolerated as evidenced by patient comfort during the procedure. This will promote better compliance for surveillance by EIS rather than mammography which requires breast compression and that be uncomfortable for some patients. Furthermore, EIS is a safe, non-invasive and fast procedure. Since there are no x-rays, repeat examinations can safely be performed. EIS is easy to use and requires minimal training for system operators and is not operator dependant in experienced hands.

The smallest lesion detected in this study measured 20 x 14 mm, which was an infiltrating ductal carcinoma. A study conducted by A. Malich et al. [9] showed that the minimum size of implanted tumour on a rabbit’s leg detected with the use of EIS is 8 mm. However, other factors such as depth and location of the lesions need to be considered. In theory, if the depth is greater than 3 cm, non-malignant tissue between the lesion and surface assimilates the electrical parameters of both tissues, and consequently the focal increased value is probably not large enough to be measured at skin surface. Even with smaller lesions, visualization depends theoretically on depth as shown by Scholz and Anderson [5].

It is of interest that, EIS in this study detected two DCIS cases. However, Malich et al [10], found significantly lower detection rates with EIS when examining noninvasive tumours (ductal carcinoma in-situ, DCIS). However, the number of DCIS cases is not sufficient in order to draw meaningful conclusions. Further studies are therefore necessary to verify our findings.

In our study we observed comparable values using EIS for “screening” of the breast as evidenced by 70% specificity in the control group. Therefore, we suggest that EIS could probably be useful for screening particularly in patients with dense breasts where conventional mammography is the least accurate. But up to now EIS does not seem to be useful as a screening tool in general especially due to the number of false positive findings.

We reported 7 false positive cases; however this may be due to artifacts from superficial skin lesions, air bubbles, interfering bones, insects bites and poor contact of probe which will decrease in frequency with increased individual experience of the examiner. In our study the sensitivity for detection of tumour is 86.6%.  The sensitivity of screening mammography (without physical examination) has been only between 85% and 90% according to a study by Bird RE [16].

In this study, 25 patients were excluded from the study because of high nipple values; of which 14 patients were diagnosed as benign proliferative lesion showing high impedance values in all 4 quadrants. All these patients show initial nipple values of more than 3.7 that gave subsequent high readings. We propose that patients with nipple values of more than 3.7, not to proceed with EIS and those other adjunctive investigations such as ultrasound or MRI should be performed instead.

Although EIS is operator independant, the ability to differentiate local brightness whether it was due to malignant lesions or artifacts from superficial skin lesions, air bubbles, interfering bones, insects bites and poor contact of probe will improve with increased individual experience of the examiner. This will subsequently reduce the false positive rate of future studies. However, some of the bright spots do not have an obvious explanation. One possibility is the influence of hormones. Changes in the level of estrogen and progesterone have been correlated with changes in dielectric properties of tissue (Fischer et al., 1990), possibly by altering the water content of tissue.

Researchers believe that the rate of false positive results with EIS seems to correlate with hormone status. In young, premenopausal women, the rate of false positives is correlated with estrogen levels over the menstrual cycle. Therefore, the best time for EIS examination in order to minimize false positive results is the second week after the start of a woman’s period. Future study with regards to this observation is needed so as to subsequently reduce the false positive rate. Similarly, the rate of false positive results is affected by taking hormone replacement therapy (HRT) after menopause. In these cases, timing of the examination does not influence results [17].

On account of its cellular properties, the nipple always shows as a bright signal. However, nipple brightness varies among patients, and perhaps in accordance to menstrual cycle and in women who taking hormone replacement therapy. Therefore, retroareolar lesions are difficult to diagnose, and in our study any change in these signals were not interpreted as a possible retroareolar lesion. Only a clearly separable focal bright spot was interpreted as a positive finding, suspicious for malignancy [5].  As such we excluded patients with retroareolar lesions due to the above limitation.

Multi-center international studies have shown that EIS is very useful in distinguishing cancers from benign lesions in BI-RADS 3 (probably benign, recommended for six month follow-up) and BI-RADS 4 (likely suspicious, recommend for biopsy). Cases with clear indications of malignancy (e.g. linear, branching, clustered punctuate or pleomorphic calcifications, masses with ill-defined or spiculated borders and clear architectural distortion) should be referred directly for biopsy [17].

CONCLUSION

EIS as an adjunct to mammography or ultrasound in classifying suspicious lesions is promising because it increases the sensitivity for cancer detection and may reduce need for biopsy of equivocal lesions.  The additional use of EIS that has a negative predictive value of 93.9% may be useful to exclude some benign lesions from further diagnostic or invasive procedures.  More studies with larger number of patients are necessary to verify these results.

The practical value of EIS depends on further reduction of false-positive results and is limited to known lesion of known location. Further studies should address whether histopathological changes associated with proliferative disease, hyperplasia or metaplasia induce an increased conductance by themselves. Advanced developments in software and scientific studies, should focus on reducing the false-positive findings in order to achieve a higher accuracy value.

Mammography is still the best method for evaluation of a breast lesion, which in this study demonstrated 100% sensitivity and 86.8% specificity. Our study showed electrical impedance scanning to have an overall sensitivity of 86.6% and specificity of 81.5%.

As a conclusion we recommend EIS in patient with equivocal breast lesions during mammography or ultrasonography especially in young patients or in patients with dense breasts provided the initial nipple values are less than 3.7 and the examination is done during the second week after menses.

REFERENCES

1.      The first report of NCR-cancer incidence in Malaysia. Report

2.      Tabar L, Fagerberg C. J. G., Gad A. Baldetrop et al. Reduction in mortality from breast cancer mass screening with mammography. Lancet 1985; 1: 829-835

3.      TransScan TS 2000 Operation Manual

4.      Jacques Jossinet. The impedivity of freshly excised human breast tissue. Physiological

5.      Measure; 19:61-75

6.      B. Scolz, Anderson. On Electrical impedance scanning- Principles and simulations. Siemens AG, Medical Engineering group, Erlangen, Germany. Electromedica 2000; 68: 35-44

7.      A. Malich, T.Boehm, M. Facius et al. Differentiation of mammographically suspicious lesion: Evaluation of breast ultrasound, MRI mammography and Electrical impedance scanning as adjunctive technologies in breast cancer detection. Clinical Radiology 2000; 56:278-283

8.      SI Fields, M. Rossmann, E. Phillips et al. Adjuctive improvement of mammographic accuracy using electrical impedance scanning. Radiology 1998; 209:272-273

9.      8.  A. Malich, T. Boehm, M. Facius et al. Use of electrical impedance scanning in the  differentiation of sonographically suspicious and highly suspicious lymph nodes of the head-neck region. European Radiology 2002;

10.  9.  Ansgar Malich, Thomas Bohm, Tobias Fritsch et al. Animal-based model to investigate the minimum tumour size detectable with an Electrical impedance scanning technique. Academic Radiology; 10:37-44

10.  A. Malich, T. Fritsch, R. Anderson et al. Electrical impedance scanning for classifying suspicious breast lesions: first results. European Radiology 2000; 10; 1555-1561

11.  A. Malich, T. Bohm, M. Facius et al. Additional value of electrical impedance scanning: experience of 240 histologically proven breast lesions. European Journal Of Cancer 2001; 37: 2324-2330

12.  Gonzalo Martin, Racio Martin, Maria Jesus Brieva et al. Electrical impedance scanning in   breast cancer imaging: correlation with mammographic and histologic diagnosis. European Radiology 2002;

13.  PJ Kneeshaw, PJ Drew and A Hubbard. Electrical impedance scanning: a new imaging technique for evaluating microcalcification in the breast? Breast cancer research 2002; 4 (Suppl): 2

14.  Fuschsjager MH, Helbich TH, Ringl H et al. Electrical impedance scanning in the differentiation of suspicious breast lesions:comparisons with mammography, ultrasound and histopathology. Rofo Fortscr Geb Rontgenstr Neuen Bilgeb Verfahr 2002 Dec; 174(12): 1522-9

15.  Jack Cuzick, Rolland Holland, and Volker Barth et al. Electropotential measurements as a new diagnostic modality for breast cancer. Lancet 1998; 352:359-363B. Scolz, Randerson. On Electrical impedance scanning- Principles and simulations. Siemens AG, Medical Engineering group, Erlangen, Germany. Electromedica 2000; 68: 35-44

16.  Bird RE. Professional quality assurance for mammography screening programs (letter) Radiology; 177:587

17.  C. Perlet, M. Kassler, S. Lenington et al. Electrical impedance measurement of the breast: effect of hormonal changes associated with the menstrual cycle. European Radiology 2000; 10:1550-1554

TABLE I: Histopathological findings, number of cases and mean size of lesions

TABLE II: Histopathological findings of all lesions with false positive EIS

TABLE III: Histopathological findings of all lesions with false negative EIS

These two cases are believed to have been within the expected range of lesions detectable by Trans Scan (one is hard scirrhous infiltrating ductal carcinoma and another one is infiltrating ductal carcinoma.

TABLE IV: Comparisons of results between mammography and electrical impedance scanning (EIS) in this study.

FIGURE I: Boxplot of level of suspicion between benign and malignant breast lesions detected using Trans Scan (conductance values). There is statistical significant difference (p<0.05) between benign and malignant breast lesions.

FIGURE II: Boxplot of level of suspicion between benign and malignant breast lesions detected using Trans Scan (capacitance values). There is statistical significant difference (p<0.05) between benign and malignant breast lesions.

(A)	(B)
(C)

FIGURE III:  (A)Mammogram shows well-defined masses in the left breast (asterisks) suggestive of benign lesions. (B)Ultrasound showed a well-defined lesion and core biopsy consistent with fibroadenoma. (C)Electrical impedance scanning images in high-resolution targeted mode shows no abnormal signals detected with low level of suspicion of malignancy (LOS less than 4).

(A)	(B)
(C)

FIGURE IV:  (A)Mammogram shows spiculated mass in the left upper outer quadrant (asterisks). (B)Ultrasound shows irregular mass with posterior acoustic shadowing that is very suggestive of malignancy and core biopsy consistent with intraductal carcinoma. (C)Electrical impedance scanning images in high-resolution targeted mode shows abnormal signals detected with high level of suspicion of malignancy (LOS more than 4).

FIGURE V: Electrical impedance scanning images in high-resolution targeted mode of the breast, from a patient in the control group. No abnormal signals were detected with low level of suspicion of malignancy (LOS less than 4) suggest negative findings.

U Table V: Performance of EIS in the detection of suspicious breast lesions

Correspondence:

Dr. Mohd Amran b. Abdul Rashid

Assistant Professor

Department of Radiology,

Kulliyyah of Medicine

International Islamic University of Malaysia

25000 Kuantan

Malaysia

Email: amranrashid@gmail.com