Clinical correlation of high activity dynamic hepatic scintigraphy in patients with colorectal cancer

Abstract

Summary The hepatic perfusion index, the ratio of hepatic arterial to total liver blood flow, was measured in 50 consecutive patients with colorectal cancer using radiolabelled colloid with high administered activity. In patients with proven liver metastases the diagnostic sensitivity of the HPI was 96% and the predictive value of a negative test was 92%. Dynamic hepatic scintigraphy is of value in the management of patients with colorectal cancer.Dynamic hepatic scintigraphy can provide an estimate of the relative proportion of hepatic arterial to total liver blood flow (the hepatic perfusion index, HPI). It has been claimed to identify up to 96% of patients with colorectal liver metastases when the HPI is abnormal. If this is correct it is to date the only technique which has the potential to identify those patients with subclinical hepatic metastases (Leveson et al., 1985;Parkin et al., 1983). Despite this, dynamic hepatic scintigraphy has not been accepted into routine clinical practice. Some investigators have failed to confirm the initially encouraging reports from Parkin et al. (Ballantyne et al., 1990), whilst others have found significant uncertainties in the curve analysis produced by low count rates (Leng et al., 1987;Goldberg et al., 1989). In an attempt to circumvent these problems we undertook a pilot study using radiolabelled colloid with a high administered activity together with image processing using Parkin's technique (Parkin et al., 1983) and have achieved a good intra-and inter-observer reproducibility and reliability of dynamic hepatic scintigraphy (Hemingway et al., 1991b). We now present our experience of dynamic hepatic scintigraphy using this modified technique in 50 consecutive patients with colorectal carcinoma. Patients and methodsFifty consecutive patients presenting with colorectal cancer underwent dynamic hepatic scintigraphy. Each patient fasted for 12 h, was positioned supine over a large field of view gamma camera and given a rapid intravenous bolus injection of 400 MBq 99'Tc albumin colloid in a volume of 2 ml via an antecubital vein. Posterior images were obtained every 2 sec for 2 min on a 64 by 64 matrix using a large field of view gamma camera (IGE 400A) fitted with a high sensitivity parallel hole collimator and interfaced to a Link Analytical MAPS computer. Regions of interest were drawn around the liver and both kidneys avoiding overlap with the lungs, aorta or spleen and time-activity curves constructed. The curves were analysed according to the method described by Perkins et al. (1987). Using the left renal artery peak (Tp) as the demarcation of the hepatic arterial and portal venous components of liver perfusion, the gradients of the 8 sec periods immediately before (G1) and after Tp (G2) were calculated using a least squares regression analysis, excluding the frame immediately overlapping Tp. The hepatic perfusion index was expressed as the ratio of the hepatic arterial gradient to the total liver blood flow gradients, i.e. HPI = GI/(GI + G2). The upper limit of nor...

Hepatic arterial chemotherapy for metastatic colorectal carcinoma

“…Methods for detecting occult hepatic metastases in colorectal cancer which are being developed include duplex sonography (Leen et al, 1991) and dynamic hepatic scintigraphy (Hemingway et al, 1992). Earlier detection leading to more rapid referral of patients with low-volume hepatic metastatic disease should result in improved response to regional chemotherapy and ultimate gain in patient survival.…”