What's new about Time Intensity Curves

Table of Contents

In daily practice contrast-enhanced ultrasound (CEUS) has proven to be helpful for detection and characterisation of liver tumours [EFSUMB-Guidelines and Good Clinical Practice Recommendations for Contrast Enhanced Ultrasound (CEUS) - Update 2008. Ultraschall in Med 2008; 29:28-44], ultrasound guided biopsy, and postablative monitoring after radio frequency ablation (RFA). With the introduction of second generation ultrasound contrast agents allowing real time imaging it is possible to describe the change of brightness over time as a function of ultrasound contrast agents in- and outflow in a certain region of interest, especially when used with current software methods allowing analysing signals of blood pool ultrasound contrast agents without disturbances of underlying tissue background information. In the literature there are many topics in which time intensity curves (TIC) may add substantial information to the information gained by conventional contrast enhanced ultrasound.

Assessment of tumour response

It is of special interest that time intensity curves may serve as a predictor of tumor response to medical treatment, e.g., to the tyrosine kinase inhibitor Imatinib [Am J Roentgenol 2006;187:1267-1273. Ultraschall in Med 2008; 5:S276-S277]. But it is also of interest that published guidelines so far have not recognized ultrasound as a recommended imaging method in the diagnosis and follow up of gastrointestinal stroma tumours (GIST). The panelists of the 2005 published guidelines agreed that currently available imaging techniques to evaluate GIST include computed tomography (CT), magnetic resonance imaging (MRI) and fluorine-18-fluoro-deoxyglucose (FDG) positron emission tomography (PET). It is also stated that the role of ultrasound is currently under investigation and CEUS was not mentioned at all [Consensus meeting for the management of gastrointestinal stromal tumors. Report of the GIST Consensus Conference of 20-21 March 2004, under the auspices of ESMO. Ann Oncol 2005;16:566]. This might be due to a so-called subjectivity of the ultrasound examination technique ("subjectivity" is certainly also true also for CT and MRI but this might be less obvious). Similarly, in the RECIST (Response Evaluation Criteria In Solid Tumors) criteria used in cancer trials only CT and MR are allowed, ultrasound is not (for more details on RECIST see http://www.eortc.be/).

Assessing hypervascularity and response to antiangiogenetic drugs

In contrast to most published reports focussing mainly on CT and PET it could be recently shown in some patients that ultrasound can be the only imaging method depicting small hypervascularized metastases and is, therefore, also the only imaging method in the follow up. In addition the early functional evaluation to optimize therapeutical strategies is a main goal of tumour evaluation using antiangiogenetic treatment. Treatment response can be predicted analyzing the vascularization before any volume reduction can be observed using the RECIST-criteria. CEUS might also be the only method identifying small lesions before and after treatment in patients with no evidence of metastases using CT, MRI and PET [Ultraschall in Med 2008; 29: S276-S277]. In an experimental animal study changes in vascularity could be detected within the first 24 hours after administration of an antiangiogenetic drug [Lavisse et al. Invest Radiol 2008; 43: 100-111]. Recognizing these patterns on CEUS in tumor response evaluation is important, since often, response in tumor size, particularly of the hepatic metastasis, is not apparent until late in therapy. The analysis of time intensity curves is, therefore, a promising method to overcome the subjectivity of contrast enhanced ultrasound.

Several of the parameters in time intensity curves correspond statistically with microvessel density, especially the area under the curve [Du et al, J Ultrasound Med 2008; 27: 821-31].

Parameters used in Time Intensity Curves

TIC-parameters used are summarised in Table [1].

Table 1 Parameters calculated from the time intensity curve and their explanation

Reproducibility

When a new technique is applied to clinical indications and questions, basic research has to be performed including the validity of the parameters analysed (intra-observer stability, stability against external factors, reproducibility) and the comparability of different software sources used.

Very few studies on the subject are available. The area under the curve may be less reproducible than other measurements; inter- and intra-observer variation of 20-30 % have been observed (Grossjohann HS, unpublished observations).

So far the amount of deviation for measuring TIC values at different depths is still remarkable. Rise time and time to peak seem to be the most stable parameters for depth positioning variations. For different sizes of the region of interest a reported mean deviation of 11 % is acceptable. For different shapes of the region of interest (circular, square, irregular) again time to peak, area under the curve and rise time are the most stable parameters when the mean deviation published is 6 %. Although there is a tendency for certain parameters (time to peak, area under the curve and rise time) to be more stable than others, all parameters had good values. It could be shown that when comparing more than one region of interest, e.g. in a tumour versus representative parenchyma they must be compared in the same depth [Ignee A, Jedrejczyk M, Schuessler G, Jakubowski W, Dietrich CF. Quantitative contrast enhanced ultrasound of the liver for time intensity curves-Reliability and potential sources of errors. Eur J Radiol 2009, epub in advance]. The size of the region of interest (ROI) could also have influence on the reproducibility of the time intensity curves; theoretically a small ROI would show larger variation than a large (but it could be shown in the mentioned study that the size of ROI does not significantly influence results). Proximity to vessels will also have influence. In patient studies it is therefore very difficult to standardize the size and the location of the ROI. Figure [1a-c ] reproducibility studies during different phases of contrast enhancement illustrating the problem of surrounding vessels.

Fig. 1a-c Reproducibility studies during different phases of contrast enhancement illustrating the problem of surrounding vessels.

Some authors stress the use of raw data files as a basic source for analysing perfusion kinetics. Conventional video data files are non-linearly compressed and discretisised on an 8 bit base with 256 grey levels. Therefore the original data are thought to be more accurate in the time intensity curve analysis. Up to now the access to raw data in commercially available ultrasound machines is limited. In addition, not all parameters are theoretically influenced by the compression like pure temporal parameters like time to peak, rise time, and mean transit time. Further investigations are necessary to define the real impact of raw data analysis in direct comparison with conventional data files.

In conclusion

Concerning the work done so far we postulate that multicentre studies are needed in order to obtain large series of patients. Before multicentre studies on strictly defined issues are performed the standardisation of time intensity curves analysis is crucial. Among the necessary topics of the standardisation are:

• the manufacturer used

• the software used

• the parameters used

• the structure of the fundamental video source - raw data or for example conventional data structures like DICOM

Time intensity curves analysis is a promising tool with plenty of potential applications. Nevertheless not much basic work has been done so far although this is important before analysing clinical questions. Please feel free and be even more encouraged to perform studies or contact us and participate on this new technology.

Cordially,

Christoph F. Dietrich,

EFSUMB Honorary Secretary

Michael Bachmann Nielsen,

Chairman EFSUMB Publications Committee