Seventy-six patients with primary gastric non-Hodgkin's lymphomas (PGL) were diagnosed, and 75 were treated between 1975 and 1985. According to the Working Formulation 22 patients had low-grade malignant histologic subtypes, 27 intermediate-grade, and 27 high-grade. Twenty-four cases were diagnosed by endoscopic biopsies, 52 through laparotomy biopsies. Forty-five underwent subtotal or total gastric resection; seven were considered unresectable at laparotomy; 23 did not undergo surgery because of the high operative risk, mainly due to advanced age and coexisting diseases; and one died of myocardial infarction a few days after admission, before starting therapy. All patients who did not undergo laparotomy were staged with bipedal lymphangiography or abdominal ultrasonography and/or computed tomography. Stage, evaluated according to the criteria of Musshoff, was I or II1 in 16 cases, II2 in five, and IV in the remaining 55. Treatment modalities included surgery (S), chemotherapy (CT), radiotherapy (RT), and combinations thereof in the following proportions: only S in ten cases, S + CT in 32 cases, S + RT in one case, S + CT + RT in two cases, CT only in 25 cases, CT + RT in five cases. No substantial differences in response to therapy and in survival were found in relation to the different treatments. Ten-year survival was 43% in Stage I or II and 20% in Stage IV. Of the 45 resected patients, five postoperative deaths were recorded (11%). No bleeding or perforations were observed in the 30 unresected patients, and survival of such cases compared with that of the resected ones. These findings, together with data from the literature, suggest that some of the advantages claimed for surgery in PGL (debulking and abatement of the risk of perforation or hemorrhage during CT or RT) have been overestimated in relation to the intrinsic surgical risk and to the possibility of anticancer therapy. Gastric resection may still be unavoidable as a diagnostic procedure in a minority of cases and may represent the primary therapeutic procedure in clinically assessed early-stage and low-risk patients, but it cannot be considered mandatory whenever possible merely for debulking purposes or to obviate possible perforation or hemorrhage. The CT and/or RT can be effective in unresected and even bulky cases, providing minimal risk of severe hemorrhage or perforation.

The aim of the study was to analyze the frequency of incidental thyroid carcinoma (unknown tumor smaller than or equal to 10 mm) in a consecutive series of 462 total thyroidectomies for multinodular goiter and to investigate the clinical risk factors for this type of malignancy. A retrospective, single-center study of outcome data collected from patients with preoperative diagnosis of multinodular goiter who underwent total thyroidectomy at the General Surgery Unit of Pavia (Italy) between January 2000 and December 2008 was performed. Possible risk factors for malignancy were: gender, age, time of evolution of goiter, presence of a dominant nodule in multinodular goiter, hyperthyroidism, history of radiation to the neck, residence in an area of endemic goiter, prior thyroid surgery, calcifications in the goiter detected by neck ultrasound or chest X-rays, and a family history of thyroid diseases. In a 9-year period, 462 patients underwent total thyroidectomy. We found 41 cases of incidental thyroid carcinoma; the most common histopathological type was papillary. The multivariable analysis demonstrated that the clinical variables associated with occult carcinoma were a personal history of radiation therapy to the neck, the presence of calcifications detected by ultrasound or neck X-rays, and a family history of thyroid diseases; residence in an area of endemic goiter was a protective factor. A personal history of radiation to the neck, detection of calcifications by ultrasound or by neck X-rays, and a family history of thyroid diseases should be considered clinical risk factors for malignancy in multinodular goiter.

Sepsis score and complement factor B (FB) have been measured in 66 severely septic surgical patients in the intensive care unit, with the aim of monitoring their clinical course and predicting their outcome. Sepsis score correlated well with clinical course. 82% of patients with initial sepsis score < 20 progressively improved and survived. Only 6% of patients with sepsis score ≧ 20 survived. FB plasma level was significantly higher (p < 0.01) in patients who subsequently survived. Two indices were identified which could predict patient outcome several days in advance with 100% accuracy: (1) the index of survival from sepsis defined as the combination of sepsis score < 20 and FB ≧ 45 mg/dl, and (2) the index of death from sepsis defined as sepsis score ≧ 20 and FB < 40 mg/dl.

No abstract