Donohue, John H.

AbstractThe etiology, prognosis, and optimal management of primary gastric carcinoids remain controversial. Records of 36 consecutive patients with gastric carcinoid (15 men) were reviewed retrospectively between 1975 and 1990. Follow‐up was complete in 97% of cases. Mean age at diagnosis was 58.4 years (range 24–82 years). The clinical presentations included anemia (72%), pain (69%), and carcinoid syndrome (11%). Associated autoimmune and endocrine abnormalities were common and included atrophic gastritis (67%), pernicious anemia (58%), hypothyroidism (39%), diabetes (19%), Addison's disease (6%), and hyperparathyroidism (6%). Lesions were nonantral in 78%, involving only the corpus in 42%, the fundus in 28%, and only the antrum in 8%; 42% were multiple. Urinary 5‐hydroxyindoleacetic acid (5‐HIAA) and serum gastrin levels were elevated in 17% and 50% of those tested, respectively. Histologic examination revealed that 28% of lesions were ≥2 cm, and 33% had liver metastases on presentation or developed them during follow‐up. Eight patients (22%) died of tumor with a median survival of 39 months. The presence of metastases, atypical histology, serosal involvement, and size > 2 cm were adverse prognostic factors. In patients without hypergastrinemia (n=6), 66% developed metastases, 60% had elevated 5‐HIAA, and 50% died of carcinoid tumor. In sharp contrast, those patients with hypergastrinemia and “typical” gastric carcinoids (n=15), metastases and death did not occur (p<0.003 and p < 0.005, respectively, compared with eugastrinemic patients). In the latter group (hypergastrinemia and typical histology), pernicious anemia was common (73%); solitary lesions and serosal penetration were seen comparatively less often (33% versus 100%, p<0.005 and 0% versus 33%, p<0.05, respectively); and no lesion measured > 2.0 cm compared to 66% in the group without elevated gastrin (p<0.002). No patient with hypergastrinaemia had elevated levels of 5‐HIAA. Overall survival at 5 and 10 years was 58% and 28%, respectively. Survival was 80% at 5 and 10 years in the hypergastrinemic group. Gastric carcinoid tumors that are histologically atypical, large (>2 cm), or present without elevated gastrin are potentially lethal and require aggressive therapy. In contrast, gastric carcinoids with typical histology and hypergastrinemia are rarely lethal, and conservative management with local excision and close observation is appropriate.

AbstractAlthough useful palliation can often be achieved when external beam irradiation and chemotherapy are used to treat locally advanced gastrointestinal malignancies, local control and long‐term survival are infrequent in view of the limited tolerance of surrounding organs and tissues. In view of dose limitations of external beam irradiation, intraoperative irradiation (IORT) with electrons has been used as a supplement to external treatment in an attempt to improve the therapeutic ratio of local control versus complications. An IORT dose of 10 to 20 Gy has been combined with fractionated external beam doses of 45 to 55 Gy in 1.8 Gy fractions in studies performed in the United States, Japan, Europe, and Scandinavian countries. In this paper the indications for and the results of aggressive combined techniques that include IORT are discussed. Results obtained with external beam techniques alone or with chemotherapy and resection are presented by site to demonstrate the need for higher doses of irradiation. When results from IORT series are compared to standard treatment with regard to disease control and survival, local control appears better with locally advanced colorectal, gastric, and pancreatic cancer; and survival appears better with colorectal ± biliary cancers. With pancreatic cancer, improvements in local control do not translate into increased survival in view of the high incidence of subsequent liver and peritoneal failures. Implications for future strategies in all sites are discussed.

Purpose Some patients with colon cancer have a high risk of local recurrence postoperatively. This trial was undertaken to determine whether radiation therapy added to an adjuvant chemotherapy regimen improves outcome in high-risk patients. Patients and Methods Patients with resected colon cancer with tumor adherence or invasion of surrounding structures, or with T3N1 or T3N2 tumors of the ascending or descending colon were randomly assigned to receive fluorouracil and levamisole therapy with or without radiation therapy. Patients who received chemotherapy and radiation therapy (chemoRT) received 45 to 50.4 Gy in 25 to 28 fractions beginning 28 days after starting chemotherapy. Patient enrollment was terminated because of slow accrual after 222 patients enrolled (original goal was 700 patients); 187 patients were assessable. Results Overall 5-year survival was 62% for chemotherapy patients and 58% for chemoRT patients (P > .50); 5-year disease-free survival was 51% for both groups (P > .50). Toxicity (≥ grade 3) occurred in 42% of chemotherapy patients and 54% of chemoRT patients (P = .04). Leukopenia (≥ grade 3) occurred in 10% of chemotherapy patients and 22% of chemoRT patients (P = .02). No significant difference in nonhematologic toxicity (≥ grade 3) was observed between chemoRT and chemotherapy patients (35% v 44%; P = .26). Conclusion Patients who received chemotherapy or chemoRT had similar overall survival and disease-free survival. Toxicity was higher among chemoRT patients. These results must be interpreted with caution because of the high number of ineligible patients and the limited power of the study to detect potentially meaningful differences.

AbstractThis prospective randomized study examined the acute‐phase response and wound healing, comparing laparoscopic (LA) and open posterior (PA) adrenalectomy in a cushingoid porcine model. Repository corticotropin gel was given to 40 pigs for 21 days. Biochemical and tissue parameters of Cushing syndrome were confirmed. The pigs were randomized to undergo LA or PA. In addition to operating time and morbidity, the acute‐phase response was compared by measuring the postoperative white blood cell count, fasting glucose, C‐reactive protein, and nitrogen balance. Wound healing was assessed by (1) scored (1–4) gross appearance at 48 hours and 1 and 2 weeks; (2) histologic examination; and (3) tensile strength. There was no difference in operating time (mean ± SD) (36 ± 9 minutes open vs. 37 ± 7 minutes laparoscopic), perioperative mortality, degree of leukocytosis, fasting glucose, or C‐reactive protein (p > 0.05). Nitrogen balance, wound scores, and tensile strength at 24 hours and 1 week were more favorable in the LA group than in the PA group (p < 0.05). In the cushingoid porcine model, laparoscopic adrenalectomy was less catabolic and was associated with fewer wound complications than the open posterior adrenalectomy. These findings provide support for continued pursuit of laparoscopic methods for adrenalectomy in the clinical setting.