Eighty consecutive individuals including 60 patients with tumor and 20 controls were enrolled in this study. Although history, physical examination, laboratory test, and Doppler sonography of liver showed that the 20 controls including 8 women at a mean age of 47.3 ± 8.6 years (range 35-64 years) had no evidence of liver disease, they underwent abdominal perfusion CT for unrelated causes. Inclusion criteria included patients with tumor confirmed by appropriate clinical and laboratory examinations, those still on chemotherapy when perfusion CT was performed, those with hepatic carcinoma or metastasis who had an adequate space in liver parenchyma for later calculation, those who had no history of alcohol abuse, viral hepatitis, liver cirrhosis or other hepatic/biliary diseases, and those who had adequate heart, renal and liver function for perfusion CT. The 60 patients including 24 women at a mean age of 53.2 ± 7.1 years (range 25-71 years) met the criteria. All patients received 1-2 cycles of chemotherapy (6-8 courses in one treatment cycle). Perfusion CT parameters were compared between controls and patients. The diagnostic criteria for hepatic steatosis were the density of liver parenchyma lower than that of spleen or the CT value of liver parenchyma less than 40 Hu on CT image without contrast agent.

The study was approved by the institutional ethics committee. Written informed consent was obtained from each patient or his or her family members after the nature of procedures was fully explained.

Perfusion CT was performed for patients at the end of each treatment cycle. After an overnight fast, patients and controls lying supine on a table underwent perfusion CT with a 64-row multi-detector CT scanner (Light Speed VCT, GE Healthcare, Chalfont St. Giles, UK). Before perfusion CT, patients underwent an abdominal scan without contrast medium during a breath hold at the end of expiration. We selected a slice with the right hepatic lobe, spleen and portal trunk clearly visualized. Then, multiple-slice dynamic sequences lasting 80 s were scanned 8 s after injection of a contrast material (iohexol, 350 mg I/mL, Changfujiejing Pharmaceutical Co. Ltd., Shandong, China). The CT perfusion protocol comprised 80 scans, for 320 images. Because blood vessels of patients with tumor during chemotherapy became too fragile to endure a high injection pressure, a bolus infusion of 40 mL contrast material was given at a speed of 2.5-3.0 mL/s via a 20-gauge intravenous catheter in the antecubital vein with a power injector (SCT-210, Medrad Inc., Indianola, PA, USA). The patients kept normal respiration during the perfusion study. The CT parameters used in this study were 120 kV, 250 mA, 5-mm slice thickness, 1-s cycle time and standard reconstruction algorithm.

After image acquisition, the data were transferred to an image processing workstation (AW4.1, GE Healthcare, Chalfont St. Giles, UK) and analyzed with the integrated software of CT Perfusion 3. Four regions of interest (ROI) were set on the abdominal aorta, portal vein trunk, spleen and right liver lobe. The ROI of right liver lobe was drawn on the whole visible right lobe carefully, avoiding blood vessels, margin of liver parenchyma and possible lesions. Hepatic perfusion index (HPI), mean transit time (MTT), and permeability-surface area product (PS) were calculated.

All data were expressed as mean ± SD. Independent-sample t test was used to determine differences between patients and controls. Data about controls and patients were compared by one-way ANOVA. Spearman correlation coefficient was used to assess the correlation between perfusion parameters, treatment cycle and ALT level. All tests were two-tailed. P < 0.05 was considered statistically significant. Data processing and analysis involved use of SPSS v11.5 (SPSS Inc., Chicago, IL, USA).

No significant difference was found in HPI (25.68% ± 7.38% vs 26.82% ± 5.13%, t = 0.77, P = 0.44), MTT (19.67 ± 5.68 s vs 21.70 ± 5.43 s, t = 1.43, P = 0.16) and PS (17.00 ± 4.56 mL/100 mL per min vs 19.92 ± 6.35 mL/100 mL per min, t = 1.90, P = 0.06) between patients and controls.

Patients were divided into two subgroups: one receiving 1 cycle of chemotherapy (n = 40) and the other receiving 2 cycles of chemotherapy (n = 20). The HPI and MTT were higher in patients receiving 2 cycles of chemotherapy than in those receiving 1 cycle of chemotherapy and controls (29.76% ± 5.87% vs 25.35% ± 4.05% and 25.68% ± 7.38%, and 25.61 ± 5.01 s vs 19.74 ± 4.54 s and 19.67 ± 5.68 s, P < 0.05). No significant difference was observed in HPI and MTT between patients receiving 1 cycle of chemotherapy and controls, and in PS between patients and controls (F = 1.78, P = 0.18) (Table 1).

Of the 60 patients, 13 (21.7%) showed hepatic steatosis. The incidence of hepatic steatosis was higher in patients receiving 2 cycles of chemotherapy than in those receiving 1 cycle of chemotherapy (P = 0.01). The HPI was higher in patients with hepatic steatosis than in those without hepatic steatosis and controls (30.85% ± 6.17% vs 25.70% ± 4.24% and 25.68% ± 7.38%, P < 0.05). No significant difference was found in HPI between patients without hepatic steatosis and controls, and in MTT or in PS between patients and controls (Table 2).

Treatment cycle was well correlated with HPI (r = 0.40, P < 0.01) and MTT (r = 0.50, P < 0.01) but not with PS (r = 0.02, P = 0.89). ALT level was not correlated with treatment cycle (r = -0.05), HPI (r = -0.06), MTT (r = 0.19), and PS (r = 0.18).

Liver damage is common in patients with tumor following chemotherapy. Although early liver damage causes no symptoms and is reversible in most patients, it occasionally progresses to more severe liver impairment, which may be irreversible, it is thus necessary to demonstrate the presence and severity of drug-related parenchymal changes.

Hepatic microcirculation changes in patients with tumor during chemotherapy were evaluated by perfusion computed tomography (CT). The hotspot of this research is whether hepatic microvascular changes can be quantified with perfusion CT and what kind of modifications can be detected.

At present, alanine aminotransferase (ALT) level is the main index in diagnosis of drug-induced hepatic damage. In this study, however, ALT level was found to be a less sensitive index and could not thoroughly reflect hepatic toxicity. The aim of this study was to investigate the microcirculation changes in liver of patients with tumor during chemotherapy, showing that hepatic perfusion index and mean transit time are significantly increased in patients undergoing 2 cycles chemotherapy.

The findings of this study may underscore the possibility of using perfusion CT parameters as indicators of hepatic microcirculation alteration in drug-induced liver damage. Perfusion CT may be used as a noninvasive tool in detection of hepatic toxicity.

The paper is a concise documentation of a good study.