Imaging investigation of liver haemodynamics in patients at risk for hepatic metastatic disease

Autor: M Blomley, Chris J Harvey
Rok vydání: 2005
Předmět:
Zdroj: The British Journal of Radiology. 78:103-104
ISSN: 1748-880X
0007-1285
DOI: 10.1259/bjr/58764378
Popis: Advances in imaging techniques have provided opportunities to explore tissue haemodynamics and study tissue microvasculature. The development of spiral CT in the 1990s facilitated single level tracking of a bolus injection of contrast medium by rapid sequential imaging of a volume of interest. Since attenuation values, measured in Hounsfield Units, are linearly related to changes in contrast agent concentration, measurement of physiological indices such as perfusion is possible. In the liver, the fact that simultaneous measurement from tissues and vessels is possible means the arterial and portal components of blood flow can be measured [1, 2]. This technique has been applied, with some success, to the study of liver haemodynamics in both diffuse liver disease [3] and in patients with colorectal carcinoma, in both overt and occult metastatic hepatic disease. In this issue Totman et al [4] have adapted the dynamic CT technique and measurements of hepatic perfusion to MR, in controls and in patients with metastatic colorectal carcinoma, with promising preliminary results. Using currently available imaging techniques metastases of 1 cm or greater can be detected with a sensitivity of approximately 100% using superparamagnetic iron oxide (SPIO) enhanced MR, and a sensitivity of 90–95% using spiral CT [5]. The sensitivity of imaging modalities for the detection of lesions smaller than 1 cm is about 50% using surgery and intraoperative ultrasound as gold standard, although the ‘‘true’’ accuracy of these standards is unknown. However, it is these subcentimetre metastases that account for disease recurrence. Clinical interest in detecting these subcentimetre liver metastases has increased recently as there is a potential role for adjuvant chemotherapy (especially given the advent of newer regimens), angiogenesis inhibitors and segmental resection, which if delivered early may alter disease progression and prognosis. Therefore functional imaging techniques have been applied in an effort to solve this problem. Using radionuclide bolus tracking studies it was shown that patients with liver metastases had a raised arterial/portal ratio (hepatic perfusion index, HPI) [6] and in patients with apparently ‘‘clear’’ livers who developed liver metastases, a significant proportion had an abnormal HPI at presentation [7]. The development of perfusion CT techniques allowed the direct measurement of separate arterial and portal hepatic perfusion using methodology modified from dynamic scintigraphy [1, 2, 8, 9]. Subsequent studies have demonstrated dynamic CT can detect perfusion abnormalities generated by occult metastases [10, 11]. Leen et al [12] developed a Doppler ultrasound perfusion index that was more sensitive than conventional imaging techniques in detecting metastases and had a strong predictive value in identifying patients who were likely to relapse. However, these results have been difficult to reproduce. Using the ultrasound microbubble contrast agent Levovist, Blomley et al [13] studied liver hepatic venous arrival time using spectral Doppler following a bolus intravenous injection. Patients with metastases demonstrated a significantly earlier arrival time and time to peak intensity than controls presumably owing to the increased hepatic arterial component as well as arteriovenous shunting in metastatic liver disease. This technique has also been shown to be a useful predictive index indicating relapse probability [14]. In addition microbubbles improved the conspicuity of metastases significantly increasing the number and smallest detectable lesion compared with baseline grey scale ultrasound and in some cases CT [15, 16]. The application of the CT methology for the derivation of hepatic perfusion to MR to investigate liver haemodynamics in metastatic colorectal carcinoma is innovative. The significantly elevated HPI in the presence of overt metastatic disease reaffirms the findings of other functional imaging modalities. This MR based approach has the obvious advantages of the lack of ionizing radiation, use of a safe contrast agent as well as operator independence. Like CT it can readily be incorporated into a clinical protocol and parametric imaging performed. The acquisition of the entire liver volume is a novel feature allowing parametric imaging to be performed at multiple levels. This is not possible with spiral CT but could be achieved using multidetector CT technology. However, in MR the relationship between signal intensity and contrast concentration is non-linear unlike CT so the absolute tissue perfusion cannot be derived in ml min ml and only a hepatic perfusion ratio can be calculated. Flow effects may also make measurement of contrast agent concentrations in vessels problematic. Also the authors do not present any data regarding the reproducibility of this technique. Dynamic CT imaging is usually performed during a single breathold whereas the MR technique is acquired during gentle breathing. The effect of this on signal changes requires further investigation. Also the length of extra time incurred in acquisition of the functional and data analysis is comparable with CT methods. Of course these preliminary results in small numbers do not allow us to draw definite conclusions regarding the robustness of Received 24 November 2004 and accepted 7 December 2004. The British Journal of Radiology, 78 (2005), 103–104 E 2005 The British Institute of Radiology DOI: 10.1259/bjr/58764378
Databáze: OpenAIRE
Externí odkaz: