Although anemia is one of the signs of colorectal cancer, the relationships between histological findings and hematological findings other than hemoglobin level have not been adequately investigated. We investigated the relationship between hematological findings, serum iron, and histological findings in 358 patients (207 men and 157 women) with colorectal cancer. Their mean (+/-SD) ages were 64.3 +/- 12.4 and 63.8 +/- 13.3 years. A hemoglobin level of less than 10 g/dl was the criterion for anemia, and 20.8% of the men and 25.8% of the women met this criterion. Univariate analysis showed that carcinoma of the cecum, ascending colon, and transverse colon; large-size carcinoma, invasion beyond the proper muscle layer; positive lymph node metastasis: and clinical stage (Dukes' B, C, and D) were factors associated with high incidence of anemia. Histological type did not affect the hematological findings. Multivariate analysis showed that age, tumor site, and tumor size were significant factors related to anemia. Depth of invasion, the presence or absence of lymph node metastasis, and Dukes' classification were not significant factors. In the presence of these factors, mean corpuscular volume and mean corpuscular hemoglobin concentration values were low, and red blood cells were microcytic and hypochromic. The incidence of a low serum iron level was about twice the frequency of a hemoglobin level of less than 10 g/dl. The results of the multivariate analysis showed that none of the factors were significantly related to iron deficiency.

No abstract

No abstract

BACKGROUND In spite of many reports, it remains unclear whether the presence of tumor cells in circulating blood flow predicts a poor prognosis. METHODS Competitive seminested reverse transcriptase‐polymerase chain reaction (RT‐PCR), a technique for the quantitative detection of tumor cells, was applied to detect the presence of tumor cells in portal and peripheral blood samples from 121 patients with colorectal carcinoma and to clarify their clinical significance. This technique can detect one carcinoembryonic antigen (CEA) mRNA‐expressing tumor cell in 1 × 105 normal lymphocytes. RESULTS Six of 33 healthy volunteers (18%) demonstrated a positive reaction to this technique. CEA mRNA expression was detected in the portal blood in 51% of patients and in the peripheral blood in 42% of patients. The results from the two blood samples were consistent in 91% of patients. The positive expression rates for portal blood in patients with T1 tumors and those with TNM Stage I disease were 38% and 45%, respectively. The positive rate was significantly higher in patients with colon carcinoma and those with Stage III or IV disease. CEA mRNA expression, quantitatively measured (× 10−8/β‐actin), was 22.9 ± 35.1 in the portal blood and 19.9 ± 40.0 in the peripheral blood, with no statistically significant difference. A significant positive correlation was noted between portal and peripheral CEA mRNA expression levels according to Speaman correlation analysis (correlation coefficient = 0.78; P < 0.01). Multivariate analysis revealed that the positive rate and level of CEA mRNA expression in the portal and peripheral blood did not appear to be influenced by the established prognostic factors. CONCLUSIONS The presence of circulating tumor cells might be of less value as a prognostic factor because they also can be detected in patients with early‐stage colorectal carcinoma and appeared to be independent of the conventional prognostic factors. Cancer 2001;92:1251–58. © 2001 American Cancer Society.

The indications for laparoscopic colon resection and the associated complications are discussed herein. This procedure was indicated for patients with benign disease or malignant disease with invasion limited to the submucosal layer. The subjects consisted of 14 cases with submucosal tumor invasion diagnosed preoperatively, three with submucosal invasion clarified by endoscopic polypectomy, three with adenomas larger than 2 cm in diameter strongly suspected of being focal submucosal tumor invasion considered unsuitable for endoscopic mucosal resection and one with Crohn's disease with ileus. Two cases in whom laparoscopic surgery was not appropriate were included in this series. In one case with a superficial elevated lesion (Ila type), 15 mm in diameter, a diagnosis of moderately differentiated adenocarcinoma of the cecum was made preoperatively, but subserosal tumor invasion of the colonic wall with negative lymph node metastasis (no) was revealed by examination of the resected specimen. The histology of the second superficial elevated lesion (Ha+ Ilc type), which had a central depression, 13 mm in diameter and located above Bauhin's valve, was a well differentiated adenocarcinoma of which the cancerous portion invaded the proper muscle with positive lymph node metastasis (n 1 ). Complications occurred in four cases. There were two cases of intraoperative vascular injury necessitating conversion to a standard laparotomy. One case with complete transection of the left ureter by End-GIA later underwent reoperation. The other case with minor leakage at the anastomotic site was managed with conservative therapy. In both of these cases the depth of tumor invasion had been incorrectly assessed as representing small elevated lesions, 15 mm in diameter, in the right colon. Furthermore, the cases who experienced complications had left colonic lesions. These results suggest that preoperative ultrasonic-endoscopy should be conducted as extensively as possible and that a good bloodless visual field appears to be necessary to avoid injuring adjacent organs. (Dig Endosc 1997 ; 9 : 1 1-1 5) words : colorectal cancer, laparoscopic colectomy, laparoscopy-assisted colectomy, laparoscopy, minimally invasive surgery Manuscript Surgery, Chelsea & Westminster Hospital, London), for their valuable advice, and would also like to thank Miss Dina Stanford for her help with the manuscript.