Contrast-enhanced ultrasound with perfusion analysis for the identification of malignant and benign tumours of the thyroid gland

Abstract

The aim of our study was to evaluate, whether the analysis of time intensity curves (TIC) of contrast enhanced ultrasound (CEUS) could help to differentiate between thyroid adenomas and carcinomas in daily clinical routine.

B-mode, Colour Coded Doppler Sonography (CCDS), Power Doppler (PD) and CEUS were applied for 50 patients (27 men, 23 women; mean age 51 years, range 16–81 years).

CEUS cine-sequences were analysed using time intensity curves (TIC) and calculating time to peak (TTP) as well as the area under the curve (AUC).

All 20 patients with carcinomas presented with a complete wash-out in the late phase of CEUS while this occurred only in three out of the 30 patients with adenomas.

Marked differences were observed between adenomas and carcinomas concerning the mean AUC in the surrounding thyroid tissue (p = 0.041). In addition, TTP differed clearly between the centre and the surrounding of the carcinomas (p < 0.05) as well as between TTP in the border area and the surrounding tissue (p = 0.01). CEUS in combination with TIC analysis allows a dynamic evaluation of the microvascularisation of thyroid nodules and is helpful for the differentiation of benign and malignant nodules.

Keywords

Thyroid carcinoma thyroid adenoma contrast-enhanced ultrasound (CEUS)perfusion imaging time-intensity-curve

1 Introduction

Ultrasound imaging is of high diagnostic significance for the identification of thyroid nodules, especially in combination with high-resolution linear probes [4]. The B-mode is thereby useful for the differentiation of cystic, complex cystic and non-cystic solid lesions being only a few millimetres in size. Colour Coded Doppler Sonography (CCDS) and Power Doppler (PD) can give additional information about the macrovascularisation of thyroid tumours [24]. However, the differentiation of benign, complex cystic lesions with necrosis and malignant tumours is challenging, especially for lesions smaller than 10 mm in size. Fine needle aspiration (FNA) is a possible approach to solve this diagnostic problem, but in case of a thyroid carcinoma a complete resection of the tumour is preferred due to the risk of metastatic spreading. Moreover, the diagnostic yield of FNA is limited to 80–99% [3 , 26] and the use of repeated FNA after a non-diagnostic result is still questionable [1].

CEUS, in particular in combination with a second generation ultrasound contrast medium and a high resolution linear probe, enables a dynamic acquisition of tumour microcirculation. The average intensity within a region of interest (ROI) can than be displayed as a function of time (time intensity curve, TIC) [8]. In terms of focal organ lesions perfusion quantification has already been proposed as an additive tool for the differentiation of benign and malignant nodules [17 , 27]. Despite these promising results, enhancement quantification has still remained challenging in daily clinical routine.

The aim of this study was to evaluate, whether CEUS with TIC analysis of the arterial and venous phase can help to differentiate histologically proven benign and malignant thyroid lesions by semiquantitatively characterising the microvascularisation.

2 Material and methods

Stored cine-sequences of examinations of the thyroid gland encompassing B-mode, Colour Coded Doppler Sonography, Power Doppler (PD) and CEUS were analysed retrospectively. The study complies with the Declaration of Helsinki. Written informed consent of all patients was obtained and the study was approved by the local ethics committee.

Ultrasound examinations of 50 patients (27 male, 23 female) between 16 and 81 years (51.8 ± 15.7 years) were analysed. Additionally, scintigraphic data for all patients were available. A 6–9 MHz multifrequency linear probe (LOGIC E9, GE Healthcare) was utilised. For the CEUS examinations 1 – 2.4 ml of sulphur hexafluoride microbubbles (SonoVue, Bracco, Italy) were injected as a bolus in one of the cubital veins followed by a saline flush of 10 ml [6]. Afterwards, the data were saved in DICOM format and stored in the local PACS system. Two independent readers analysed the dynamic loops of the arterial phase (15 – 45 s) and the dynamic effects in the late phase (120 – 180 s) using consensus decision. The TIC-analyses for the determination of the TTP and the AUC were performed at the ultrasound machine itself. For this purpose, ROIs 5 mm in size were drawn in the centre of the lesion (two ROIs), in the border area of the lesion (3, 6, 9 and 12 o’clock) and in the surrounding thyroid tissue (two ROIs). If necessary, motion correction was applied. The wash-out was evaluated via independent monitoring of the late contrast phase (70 – 180 s). The ROIs within the tumours were afterwards compared to the ROIs of the peripheral thyroid tissue concerning echogenicity (B-mode) and macrovascularisation (CCDS). Criteria of malignancy were a low or inhomogeneous echo at B-mode, an irregular tumour configuration, ill-defined borders of the tumour and signs of irregular hypervascularisation in CCDS. The diagnosis of an adenoma was made when the tumour had a well-defined border, a dark rim at B-mode and prominent vessels at the margin. Concerning CEUS an irregular hypervascularisation as well as a wash-out in the late phase were suspect for malignancy. In case of complex cystic lesions attention was payed to septal microvascularisation, a nodular enhancement and necrotic areas within the lesions.

In all cases surgery was performed within two weeks after the ultrasound examination.

2.1 Statistics

For statistic evaluation PASW 18 (PASW v. 18, IBM SPSS, Inc., Armonk, NY, USA) was used. Statistical analyses were performed using the Mann-Whitney-U-, the Wilcoxon-W- and the Scheffé-test. A p-value of <0.05 was considered statistically significant.

3 Results

Intolerance to the contrast medium did not occur in any patient. For three data sets a motion correction was necessary due to marked motion artefacts.

Histopathology proved 20 malignant lesions (three follicular, 13 papillary and four medullary carcinomas) and 30 adenomas. Six of the adenomas had a complex cystic structure.

At B-mode 13 malignant lesions showed an irregular margin, 18 a low inhomogenous echo.

Adenomas had a well-defined margin in 22 cases and signs of hypervascularisation at the edges at CCDS in 20 cases (Fig. 1).

CEUS revealed an irregular hypervascularisation of 16 malignant lesions with a complete wash-out for all 20 malignant lesions (Fig. 3). Adenomas, on the contrary, showed a wash-out in only three cases.

In four of the complex cystic adenomas a partial hypervascularisation was already observed at CCDS. After contrast medium administration all of the complex cystic lesions showed an irregular inner cystic enhancement (Fig. 2).

The mean TTPs for carcinomas were 10.4 ± 3.7 s in the centre (range 4.3–20.9 s), 11.2 ± 4.1 s in the border area of the lesion (range 5.8–23.3 s) and 13.8 ± 6.9 s in the surrounding thyroid tissue (range 5.2–29.9 s).

For adenomas the mean TTPs were 10.0 ± 3.2 s (range 5.8–16.3 s) in the centre, 9.9 ± 3.4 s (range 6.3–15.0 s) in the border area of the lesion and 10.7 ± 4.0 s (range 6.8–16.2 s) in the surrounding thyroid tissue.

For the complex cystic adenomas the mean TTPs were 10.1 ± 5.4 s (range 6.7–21.8 s) in the centre, 9.3 ± 2.1 s (range 7.1–12.7 s) in the border area and 10.3 ± 4.3 s (range 7.8–19.7 s) in the surrounding thyroid tissue (Table 1).

Marked differences were observed between adenomas and carcinomas concerning the AUC in the surrounding thyroid tissue (p = 0.041). The difference between the TTP of the centre and the boarder area showed a clear trend (p = 0.067).

For carcinomas the analysis of the AUC revealed mean values of 265.4 ± 307.8 (range 39.9–1328.8) in the centre, 283.9 ± 281.8 (range 62.8–1243.7) in the border area of the lesion and 152.8 ± 180.8 (range 5.9–655.5) in the surrounding thyroid tissue.

For adenomas the mean AUC values were 333.6 ± 592.0 (range 49.3–1006.5) in the centre, 338.9 ± 222.7 (range 89.5–1133.2) in the border area of the lesion and 288.9 ± 171.3 (range 91.6–881.3) in the surrounding thyroid tissue.

In case of complex cystic adenomas the mean AUC values were 78.9 ± 79.0 (range 16.7–251.4) in the centre, 220.6 ± 126.0 (range 51.6–416.1) in the border area of the lesion and 174.8 ± 75.5 (range 89.6–269.5) in the surrounding thyroid tissue (Table 2).

For all 50 tumours a statistically significance for AUCs between the surrounding thyroid tissue and the centre (p = 0.004) as well as for the surrounding thyroid tissue and the border area (p < 0.0005) could be observed. Especially for carcinomas the differences between the AUCs of the surrounding thyroid tissue and the boarder area and between the surrounding thyroid tissue and the centre were remarkably high (p < 0.0005). For adenomas a significant difference between the AUCs of the surrounding thyroid tissue und the border area (p = 0.014) was observed.

Using a cut-off value of 270.0 for the mean AUC value in the centre of the tumour a sensitivity of 80% and specificity of 63% for the differentiation of malignant lesions and adenomas can be achieved.

Additionally, the Scheffé tests showed that the mean AUC values for the surrounding tissue of carcinomas and adenomas differed significantly. A higher mean AUC value in the surrounding tissue of a tumour makes an adenoma thereby more probable than a carcinoma.

4 Discussion

Contrast-enhanced ultrasound enables a dynamic evaluation of the microvascularisation of thyroid tumours. Initially, CEUS was established for the characterisation of benign and malignant liver lesions [8 –11]. Multicentre studies revealed that CEUS achieved the same diagnostic reliability as contrast-enhanced computer tomography (ceCT) and contrast-enhanced magnet resonance imaging (ceMRI)[5 , 23]. Contrast enhanced harmonic imaging (CHI) together with the new second generation contrast agents allows now the continuous acquisition of tumour vascularisation up to the capillary level. Starting only 15 seconds after contrast medium application the perfusion pattern of the tumour can be visualised from the early arterial up to the late phase (3–5 min). Using Contrast Harmonic Imaging, mostly as amplitude or phase emission technique at a low mechanical index (MI <0.2), the 3–10μm thick microbubbles start to oscillate. These oscillations lead to an amplification of the echo signal up to 20 dB. Together with the use of a high resolution linear probe with transmission sequences up to 15 MHz a high degree of details can be achieved [22].

In line with the current non-liver guidelines no final assessment about thyroid tumours concerning the diagnostic accuracy of CEUS exists [16]. CEUS provides information about central or peripheral hypervascularisation, the existance of vascularised cystic or necrotic areas and the degree of wash out of the nodule in comparison to the surrounding tissue. In our study, a complete wash-out in the late phase was observed in all carcinomas but only in three of the adenomas.

In addition to the visualisation of characteristic enhancement patterns the results implicated that parameters like TTP and AUC can improve the differentiability between malignant and benign thyroid nodules. A higher mean TTP in the surrounding thyroid tissue compared to the mean TTP in the centre of the nodule hints towards a malignant process. In a recent study, Jiang et al. examined 213 thyroid nodules with CEUS, comparing quantitative parameters as maximum intensity of peak (IMAX), rise time from 10% to 90% of IMAX and TTP for benign and malignant nodules. In line with our results, the authors came to the conclusion that a quantitative evaluation is helpful in differentiating between benign and malign thyroid nodules [7].

Although a rapid TTP might be expected in the centre of malignant nodules, our study resulted a significant difference between the mean TTPs in the centre and the border area of adenomas and carcinomas. Similar results have already been published by other groups, underlining the overlap in the vascularisation of malignant and benign nodules [2, 15]. This could be due to the reduced microvascularisation in carcinomas as a result of necrosis. This assumption is supported by a recent study with Doppler ultrasound, reporting absent vascularity in 60% of malignant nodules, compared with 43% of benign nodules [13]. Notably, we found a significant difference between the mean TTPs in the border area of the carcinomas and the surrounding thyroid tissue and between the mean TTPs in the centre and the surrounding thyroid tissue. Moreover, high AUC values in the surrounding thyroid tissue of tumours seemed to hint towards adenomas whereas carcinomas seemed to be associated with lower AUC values of the surrounding thyroid tissues.

Independent from the quantitative analyses visual analyses of the shape of all TIC-curves pointed out that a rapid and high level of enhancement in the arterial phase together with a complete wash-out in the late venous phase indicated a malignant process. A high level of enhancement and a moderate increase of the wash-in curve combined with a delayed wash-out hinted towards an adenoma. Complex cystic adenomas had similar curve shapes than adenomas whereat they did not show a comparable intensive contrast enhancement than solid adenomas (Fig. 4).

Because multicentre studies about CEUS and thyroid lesions are missing, histopathology should still be the gold standard. However, a quantitative analysis of nodular microvascularisation using CEUS can provide helpful additional, but not definite information when examine an obscure thyroid nodule. Scintigraphy allows the assessment of thyroid nodular function, but its diagnostic specificity decreases in small lesions (<10 mm) [12, 20].

4.1 Limitations of the study

One limitation of CEUS is based on its bolus kinetic, which is associated with the restriction that only selected lesions can be assessed. A continuous kinetic up to three minutes would require an application of contrast medium boli for every single lesion. Therefore, an experienced examiner is necessary to detect and evaluate suspect lesions in the early phase. If a second injection of contrast medium is required a prolonged waiting period of at least 30 minutes is a prerequisite. Recently introduced colour-coded perfusion programs allow now the dynamic evaluation of several lesions with one single contrast agent bolus by using FLASH kinetic analysis [21].

Another limitation lies in the limited sample size, possibly resulting in the absence of a significant difference between some parameters. Therefore, larger cohorts should be analysed to improve the validity of our initial observations.

5 Conclusion

The study implicates that CEUS in combination with TIC analysis allows a dynamic evaluation of the microvascularisation of indistinct thyroid nodules in daily clinical practice, thereby facilitating the preoperative sonographic accuracy. Low values for AUC in the surrounding thyroid tissue, an accelerated TTP in the centre of a thyroid nodule in contrast to its surrounding tissue and a complete wash-out in the late phase are strong indication for a thyroid carcinoma.

References

Anderson

T.J.T.

, Atalay

M.K.

, Grand

D.J.

, Baird

G.L.

, Cronan

J.J.

and Beland

M.D.

, Management of Nodules with Initially Nondiagnostic Results of Thyroid Fine-Needle Aspiration: Can We Avoid Repeat Biopsy? Radiology 272(3) (2014), 777–784.

Appetecchia

, Bacaro

, Brigida

, Milardi

, Bianchi

and Solivetti

, Second generation ultrasonographic contrast agents in the diagnosis of neoplastic thyroid nodules, J Exp Clin Cancer Res 25(3) (2006), 325–330.

Ceresini

, Corcione

and Morganti

et al., Ultrasound-guided fine-needle capillary biopsy of thyroid nodules, coupled with on-site cytologic review, improves results, Thyroid 14(5) (2004), 385–389.

Cooper

D.S.

, Doherty

G.M.

and Haugen

B.R.

et al., Management guidelines for patients with thyroid nodules and differentiated thyroid cancer, Thyroid 16(2) (2006), 109–142.

Egger

, Goertz

R.S.

and Strobel

et al., Dynamic contrast-enhanced ultrasound (DCE-US) for easy and rapid evaluation of hepatocellular carcinoma compared to dynamic contrast-enhanced computed tomography (DCE-CT)–a pilot study, Ultraschall in Med 33(6) (2012), 587–592.

Greis

, Technical aspects of contrast-enhanced ultrasound (CEUS) examinations: Tips and tricks, Clin Hemhorheol Microcirc 58(1) (2014), 89–95.

Jiang

, Huang

and Zhang

et al., Contrast-enhanced sonography of thyroid nodules, Clin Ultrasound JCU (2014).

Jung

E.M.

, Clevert

D.A.

and Schreyer

A.G.

et al., Evaluation of quantitative contrast harmonic imaging to assess malignancy of liver tumors: A prospective controlled two-center study, World J Gastroenterol 13(47) (2007), 6356–6364.

Kim

T.K.

and Jang

H-J

, Contrast-enhanced ultrasound in the diagnosis of nodules in liver cirrhosis, World J Gastroenterol 20(13) (2014), 3590–3596.

10.

Larsen

L.P.S.

, Rosenkilde

and Christensen

et al., The value of contrast enhanced ultrasonography in detection of liver metastases from colorectal cancer: A prospective double-blinded study, Eur J Radiol 62(2) (2007), 302–307.

11.

, Guo

and Hua

et al., Characterization of focal liver lesions: Comparison of pulse-inversion harmonic contrast-enhanced sonography with contrast-enhanced CT, Clin Ultrasound: JCU 35(3) (2007), 109–117.

12.

Meier

D.A.

, Kaplan

M.M.

, Radioiodine uptake and thyroid scintiscanning.291– 313, viii, Endocrinol Metab Clin North Am 30(2) (2001).

13.

Moon

H.J.

, Kwak

J.Y.

, Kim

M.J.

, Son

E.J.

and Kim

E.-K.

, Can vascularity at power Doppler US help predict thyroid malignancy? Radiology 255(1) (2010), 260–269.

14.

Nayar

and Ivanovic

, The indeterminate thyroid fine-needle aspiration: Experience from an academic center using terminology similar to that proposed in the National Cancer Institute Thyroid Fine Needle Aspiration State of the Science Conference, Cancer 117(3) (2009), 195–202.

15.

Nemec

, Nemec

S.F.

, Novotny

, Weber

, Czerny

and Krestan

C.R.

, Quantitative evaluation of contrast-enhanced ultrasound after intravenous administration of a microbubble contrast agent for differentiation of benign and malignant thyroid nodules: Assessment of diagnostic accuracy, Eur Radiol 22(6) (2012), 1357–1365.

16.

Piscaglia

, Nolsøe

, and Dietrich

C.F.

, et al., The EFSUMB Guidelines and Recommendations on the Clinical Practice of Contrast Enhanced Ultrasound (CEUS): Update on non-hepatic applications, Ultraschall in Med 33(1) (2012), 33–59.

17.

Rognin

N.G.

, Arditi

and Mercier

et al., Parametric imaging for characterizing focal liver lesions in contrast-enhanced ultrasound, IEEE Trans Ultrason Ferroelectr Freq Control 57(11) (2010), 2503–2511.

18.

Seitz

, Strobel

and Bernatik

et al., Contrast-Enhanced Ultrasound (CEUS) for the characterization of focal liver lesions - prospective comparison in clinical practice: CEUS vs. CT (DEGUM multicenter trial). Parts of this manuscript were presented at the Ultrasound Dreiländertreffen Davos, Ultraschall in Med 30(4) (2009), 383–389.

19.

Testa

A.C.

, Timmerman

and Van Belle

, et al., Intravenous contrast ultrasound examination using contrast-tuned imaging (CnTI) and the contrast medium SonoVue for discrimination between benign and malignant adnexal masses with solid components, Ultrasound Obstet Gynecol 34(6) (2009), 699–710.

20.

Tollin

S.R.

, Fallon

E.F.

, Mikhail

, Goldstein

and Yung

, The utility of thyroid nuclear imaging and other studies in the detection and treatment of underlying thyroid abnormalities in patients with endogenous subclinical thyrotoxicosis, Clin Nucl Med 25(5) (2000), 341–347.

21.

Tranquart

, Mercier

, Frinking

, Gaud

and Arditi

, Perfusion quantification in contrast-enhanced ultrasound (CEUS)–ready for research projects and routine clinical use, Ultraschall in Med 33(Suppl 1) (2012), S31–S38.

22.

Wege

A.K.

, Schardt

, Schaefer

, Kroemer

, Brockhoff

and Jung

E.M.

, High resolution ultrasound including elastography and contrast-enhanced ultrasound (CEUS) for early detection and characterization of liver lesions in the humanized tumor mouse model, Clin Hemorheol Microcirc 52(2-4) (2012), 93–106.

23.

Wendl

C.M.

, Müller

, Meier

, et al. High resolution contrast-enhanced ultrasound and 3-tesla dynamic contrast-enhanced magnetic resonance imaging for the preoperative characterization of cervical lymph nodes: First results, Clin Hemorheol Microcirc 52(2-4) (2012), 153–166.

24.

M.-H.

, Chen

C.-N.

and Chen

K.-Y.

et al., Quantitative analysis of dynamic power Doppler sonograms for patients with thyroid nodules, Ultrasound Med Biol 39(9) (2013), 1543–1551.

25.

Xiang

, Hong

and Zhang

et al., Contrast-enhanced ultrasound (CEUS) facilitated US in detecting lateral neck lymph node metastasis of thyroid cancer patients: Diagnosis value and enhancement patterns of malignant lymph nodes, Eur Radiol 24(10) (2014), 2513–2519.

26.

Yassa

, Cibas

E.S.

and Benson

C.B.

et al., Long-term assessment of a multidisciplinary approach to thyroid nodule diagnostic evaluation, Cancer 111(6) (2007), 508–516.

27.

, Han

and Chen

, Contrast-enhanced ultrasound for differentiation of benign and malignant thyroid lesions: Meta-analysis, Otolaryngol Head Neck Surg 151(6) (2014), 909–915.