Gastroaortic fistula as an early complication of esophagectomy.

Gastroaortic fistula following esophagogastrectomy is an uncommon and invariably fatal complication without urgent surgical intervention. We report 1 such case and review the world literature identifying 22 previous cases. It characteristically presents 2 to 3 weeks after esophagogastrectomy with an initial herald bleed, followed by a latent period with a mean duration of 10 hours (range: 30 minutes to 3 days) and final exsanguination. Only 1 patient in this series survived. Awareness is necessary to allow prompt diagnosis and treatment of this almost invariably lethal but curable condition.

Regional indices of relative hepatic arterial perfusion from dynamic liver scintigraphy: the variability of indices and the use of parametric imaging.

Regional indices of relative arterial hepatic perfusion have been studied in 21 control subjects following dynamic radiocolloid scintigraphy of the liver (DLS). Three different indices have been calculated: the hepatic perfusion index (HPI); the hepatic arterial ratio (HAR) and the mesenteric fraction (MF). Three regions were defined in the upper, mid and lower right hepatic lobes and the three indices were calculated for each region. There was reasonable agreement between regional values of the same index with inter-regional correlation coefficients above 0.7 and standard errors in straight line fits of less than 0.093. There were significant regional differences for (1-MF) and HPI indices, but not for HAR. The index (1-MF) was calculated for each pixel and presented as a parametric image in 16 control subjects. The parametric images indicated the raised regional arterial indices due to overlying lung, right kidney and aorta. Parametric images may be of value to show the hepatic area free of significant overlying tissue and therefore available for analysis by DLS. However, the results suggest that the observed variability of single pixel indices limit the potential of parametric imaging for the localization of small focal lesions.

A comparison of methods to measure the percentage hepatic replacement with colorectal metastases.

The percentage hepatic replacement (PHR) with liver metastases has been recognised as an important factor for patient management, prognosis, and clinical trial evaluation. The PHR can be assessed by various methods including ultrasound, radioisotope, or computed tomography (CT) imaging, or by inspection and palpation at laparotomy. We report a prospective study comparing these methods of PHR assessment in 56 consecutive patients with colorectal liver metastases. When all four methods were used complete agreement was achieved in only 35% of cases (9/26). The three imaging techniques agreed in only 37% of cases (14/38). Overall, the correlation between different methods was poor with a tendency for surgery to underestimate the PHR. However, CT and surgery agreed in 81% of cases (21/26), and CT gave a higher estimate than surgery in each of the other five patients. With such variability in the estimates of PHR derived from these different techniques, consideration must be given to which technique is used. It is concluded that CT is probably the most accurate for PHR assessment at present.

Pre-operative scanning of the liver for colorectal liver metastases.

The comparative value of staging the extent of colorectal liver metastases by isotope, ultrasound and CT imaging has been assessed in 20 patients considered for inclusion into a randomized trial. Metastases were identified in all patients by CT, but failed in 4 by isotope and 2 by ultrasound scanning. There was often a discrepancy of 25% in estimating tumour replacement between the different modalities and surgical evaluation. As the extent of tumour is an important prognostic factor, initial evaluation should incorporate all three imaging techniques. Relative regional blood flow in tumour and normal liver regions was measured by dynamic liver scintigraphy, showing that 60% have a predominantly arterial blood supply.

Temporary blood flow stasis with degradable starch microspheres (DSM) for liver metastases in a rat model.

A reliable liver metastasis model using intraportal injections of sarcoma cells was established in syngeneic hooded Lister rats to study the blood supply of the tumours and to evaluate the role of degradable starch microspheres (DSM) in conjunction with selective hepatic arterial and portal venous chemotherapy. The tumour/normal liver (T/L) ratio after intra-arterial and intraportal injection of 113Sn microspheres was 1.04 (range: 0.38-1.15) and 0.03 (range: 0.006-0.22), respectively. After intravenous 14C-iodoantipyrine quantitative autoradiography of tumour and normal regions demonstrated a mean T/L ratio of 0.74 +/- 0.05. After hepatic artery ligation (HAL) and portal vein ligation (PVL) the values were 0.32 +/- 0.05 and 0.42 +/- 0.05, respectively. These results confirm that the vascularity of the tumours in this model is similar to human colorectal cancer metastases. Radiolabelled 14C 5-Fluorouracil (5-FU) was given intravenously, via the hepatic artery and via the portal vein (the latter two routes with and without DSM). Quantitative autoradiography of tumour regions showed that selective hepatic arterial administration with DSM resulted in a significantly increased concentration of 14C-5-FU within the tumours. These results suggest that DSM may enhance the therapeutic benefit of hepatic arterial 5-FU by increasing its uptake into tumours.

An association between parameters of liver blood flow and percentage hepatic replacement with tumour.

The extent of hepatic replacement with tumour is a significant prognostic factor in patients with liver metastases. Measuring the percentage hepatic replacement (PHR) accurately is difficult, but is important for both patient management and clinical trial evaluation. This study correlates haemodynamic indices obtained by dynamic liver scintigraphy (DLS) with estimates of PHR made from isotope scan, ultrasound, CT scan and laparotomy in 45 patients with established colorectal liver metastases and 21 controls who also underwent DLS. There was a significant reduction in the mesenteric fraction (MF) in the group of patients with metastases compared to the controls (P less than 0.001), and also a significant trend for progressive reduction in the MF with increasing PHR. A significant rise in an index of total hepatic arterial blood flow was also demonstrated with increasing PHR. These results are important with current interest in regional hepatic arterial therapy, and may prove of clinical value for prediction or monitoring of response to therapy.

Prospective randomized controlled trial of hepatic arterial embolization or infusion chemotherapy with 5-fluorouracil and degradable starch microspheres for colorectal liver metastases.

Survival benefit from hepatic artery embolization (HAE) or hepatic arterial infusion chemotherapy (HAI) in patients with unresectable colorectal liver metastases has not previously been assessed in a randomized controlled trial. Sixty-one patients were randomized, 20 to receive no treatment, 22 to receive HAE, and 19 to receive HAI with 5-fluorouracil and degradable starch microspheres. Both treatments were acceptable to the patients in terms of low treatment morbidity rate. Median survival from diagnosis of metastases was 9.6 months for controls, 8.7 months for the HAE group and 13.0 months in the HAI group. There was no apparent survival benefit for the HAE group. The increased survival in the HAI group was observed in all the subgroups analysed but failed to reach statistical significance. The greatest observed benefit was achieved in the subgroup with less than 50 per cent hepatic replacement with tumour at presentation (median survival from diagnosis 10.0 months for controls, 10.2 months for HAE and 23.6 months for HAI); 36 per cent of patients developed extrahepatic disease recurrence. No significant benefit has been shown from either HAE or HAI, but a more carefully selected group of patients with only low volume hepatic disease may benefit from HAI therapy.

Liver tumour blood flow and responses to arterial embolization measured by dynamic hepatic scintigraphy.

Liver and tumour blood flow has been studied in 30 patients with multiple liver metastases and in 14 patients with solitary liver tumours by means of dynamic hepatic scintigraphy. Observations were compared with those of a group of 33 control subjects. Haemodynamic changes were also measured in 10 patients who underwent hepatic arterial embolization (HAE). The mesenteric fraction (MF) to tumour regions in 32 subjects showed a wide range compared with control subjects. In 9 patients the MF to the tumour region was within the normal range suggesting that some tumours may possess a portal venous supply. The MF to the uninvolved liver regions was below the normal range in 25% of patients, indicating that HAE could be hazardous in this group. Following HAE the MF rose in all 4 tumour regions and fell in 4 non-embolized uninvolved liver regions. No increase in colloid clearance rate (k) was seen though a significant decrease occurred in 4 patients. These changes may well represent increased portal venous flow into tumours.

A comparison of three indices of relative hepatic perfusion derived from dynamic liver scintigraphy.

Data from dynamic radiocolloid liver scintigraphy (DLS) have been analysed to calculate three indices of relative arterial to total hepatic perfusion. Ninety subjects have been studied, comprising 21 normals, 62 patients with metastatic liver disease and 7 patients with cirrhosis. Correlation coefficients above 0.81 were found in all patient groups between an index based on rates of liver uptake (the hepatic perfusion index, HPI) and a method based on quantitative liver uptake (the mesenteric fraction, MF). A further method employing the spleen to model arterial inflow (hepatic arterial ratio, HAR) had less agreement with both HPI and MF, with correlation coefficients below 0.76. Posterior images have previously been used to calculate HAR, and greater errors are expected in HAR from the anterior images acquired in this study. Receiver operating characteristic analysis showed that the diagnostic performance of HPI and MF indices in metastatic disease were not significantly different. For anterior image data analysis both HPI and MF were superior to HAR.