From January, 1972 to June, 1989, 51 patients with liver hemangiomas (32 females and 19 males, mean age 35 years) were evaluated for surgical treatment. Diameters of the masses were 5 cm to 20 cm (median 8.5 cm). Nine of the patients had already been treated for cancer. Twenty-two (43.1%) of the 51 patients were symptomatic and 29 (56.9%) patients were asymptomatic. In 34 patients (66.7%) a definite diagnosis of hemangioma was made by scintiscan and/or ultrasound and/or computed tomography and/or angiography while in the remaining 17 (33.3%) patients the diagnosis was uncertain. The most common indications for resection were the presence of a symptomatic angioma, a symptomatic mass with an uncertain diagnosis, and/or lack of a definite pre-operative diagnosis. Surgery was performed on 25 patients. Ten anatomic and 15 atypical resections or enucleations were performed. There were no postoperative deaths. Two further patients, operated for probable hemangioma, were found to have primary hepatic malignancies. In the 26 unresected patients, no complications were observed during follow-up. In 3 patients, hemangioma enlargement was detected by ultrasound, but there were no symptoms. As cavernous liver hemangiomas are now more reliably diagnosed and their natural history is usually uneventful, surgery can be avoided in most cases. However, when a non-resection policy is adopted, an exact diagnosis is essential in order to rule out primary or metastatic cancer. Surgical exploration and treatment should be limited to symptomatic or complicated cases as well as to patients with an uncertain diagnosis.

In recent years, wide agreement has been expressed on the value of surgical resection for liver metastases from colorectal cancer, while for unresectable patients different types of locoregional treatment have been attempted. One hundred seventy-one patients with hepatic metastases from colorectal cancer were treated by us over a period of 15 years. Sixty-four underwent hepatic resection, and 107 underwent various forms of locoregional treatment. Our experience confirms the opinion that hepatic resection can be performed with a "curative" aim in patients with colorectal liver metastases; a 5-year survival rate can be achieved in about 30% of resectable cases. Adjuvant chemotherapy after hepatic resection should be tested in prospective randomized trials. Patients with diffuse liver metastases can benefit from locoregional infusion chemotherapy. Symptoms improve in most patients and objective responses are higher than those reported for systemic chemotherapy. Survival benefit with respect to untreated patients, has not yet been demonstrated by prospective randomized studies. Future improvements may be achieved by using new treatment modalities, such as new drug combinations, repeat arterial ischemia, and local tumor destruction. As these types of treatment are still experimental they should be employed only in prospective clinical trials.

No abstract

Transmucosal gastric potential difference (TGPD) was measured in the antrum and fundus of the stomach in two groups of rats submitted to hemorrhagic shock. In the first group the stomach contained 2 cm³ of 0.1 N HCl and in the second 2 cm³ of physiological saline. After the hemorrhage both antral and fundal TGPD diminished significantly in both groups. Antral TGPD dropped from –20 to –6 mV (p < 0.001) in the first group and from –22 to –12 mV (p < 0.01) in the second group; fundal TGPD dropped from –41 to –16 mV (p < 0.001) in the first group and from –40 to –17 mV (p < 0.05) in the second group. 20 min after reinfusion of blood extracted during the hemorrhage, both antral and fundal TGPD returned to normal values in the rats instilled with physiological saline, while in those treated with HCl TGPD values remained at levels significantly lower than the baseline values (in the antrum –10 mV, p < 0.001; in the fundus –25 mV, p < 0.02). Only those rats whose stomachs contained HCl developed ulcers, mainly located in the fundus of the stomach. These results suggest that the energy metabolism of the cells of the gastric mucosa undergoes constant alteration from the earliest stages of hemorrhagic shock. These alterations are greater in the fundus than in the antrum, a fact compatible with the greater incidence of ulcers in the fundus. It is probable that in the presence of HCl in the stomach, back-diffusion of H⁺ occurs, leading to the formation of ulcers, while the metabolic alterations occurring in the absence of acid proved to be reversible.

This study clearly shows that cisplatin intrahepatic arterial infusion is able to provide a good palliative effect with a tolerable toxicity.