Endoscopic injection therapy in bleeding Mallory-Weiss syndrome: a randomized controlled trial

BACKGROUND

Endoscopic injection is widely used in the therapy of bleeding gastroduodenal ulcers, but its role in the management of bleeding Mallory-Weiss tears has not been properly assessed.

METHODS

Sixty-three patients undergoing emergency endoscopy in whom there was a high index of suspicion that a Mallory-Weiss tear was the source of bleeding were randomly assigned to undergo endoscopic injection therapy (epinephrine and polidocanol) or no endoscopic therapy in 2 university-affiliated hospitals. Rates of recurrent bleeding, transfusion requirements, complications, mortality, and length of hospital stay were determined for both groups of patients.

RESULTS

Bleeding recurred in 8 patients in the control group versus only 2 in the endoscopic treatment group (25.8% vs. 6.2%, p < 0.05). Hospital stay was longer for the control group (5.5 +/- 0.2, median 6.0, range 2.0-8.0 days vs. 3.4 +/- 0.2, median 3.0, range 2.0-6.0 days; p < 0.001). There was a trend toward a higher transfusion requirement after endoscopy in the control group versus the patients treated by injection (0.9 +/- 0.2, median 0.0, range 0.0-4.0 units vs. 0.2 +/- 0.1, median 0.0, range 0.0-2.0 units; p = 0.09). No complications or adverse events caused by endoscopic injection were noted. Two patients in the control group died of causes unrelated to bleeding.

CONCLUSIONS

Endoscopic injection therapy is a useful option in the management of patients with Mallory-Weiss syndrome at high risk for recurrent bleeding.

[Hepatocellular carcinoma in a patient with hereditary hemochromatosis without cirrhosis]

Hepatocellular carcinoma in patients with hereditary hemochromatosis in the cirrhotic phase is one of the complications causing greatest mortality and may present in spite of removal of excess iron by bloodletting. Hepatocellular carcinoma is usually considered to occur in cirrhotic livers and consequently measures for the early diagnosis of this complication are only recommended in this type of patient. We present the case of a 69-year-old female patient with non-cirrhotic hemochromatosis who, 6 years after undergoing successful treatment, developed hepatocellular carcinoma. This observation should be added to the 12 cases published in the literature. Criteria should be established for the early diagnosis of hepatocellular carcinoma in patients with hereditary hemochromatosis, irrespective of whether they have cirrhosis.

Laparoscopic-assisted vs. open colectomy for colorectal cancer: influence on neoplastic cell mobilization

Laparoscopic surgery for treatment of colorectal cancer has been suggested to enhance tumor dissemination. Recently, molecular techniques have been developed to detect micrometastatic disease in patients with solid tumors, with a higher accuracy than cytologic or immunohistochemical approaches. This study was undertaken to investigate the potential harmful effects of laparoscopic-assisted colectomy on neoplastic cell mobilization in patients with resectable colorectal cancer. Fifty patients with nonmetastatic colorectal cancer were randomly assigned to laparoscopic-assisted (LAC, n = 26) or open (OC, n = 24) colectomy. Peripheral venous blood samples were obtained preoperatively, immediately after tumor removal, and 24 hours later. In 10 patients from each treatment group, portal blood and peritoneal fluid samples were also obtained before and after resection. Neoplastic cells were detected by means of reverse transcriptase-polymerase chain reaction targeted to carcinoembryonic antigen (CEA) transcription. CEA mRNA was detected in peripheral venous blood samples from 35 of 50 colorectal cancer patients preoperatively. Among those 15 baseline-negative patients, four experienced conversion 24 hours after tumor resection (2 [33%] of 6 in the LAC group vs. 2 [22%] of 9 in the OC group; NS). At that time point, clearance of CEA mRNA expression was observed in 14 of the 35 baseline-positive patients (9 [45%] of 20 in the LAC group vs. 5 [33%] of 15 in the OC group; NS). In addition, only one patient in the LAC group with baseline-negative CEA mRNA expression experienced portal blood conversion after tumor removal, although his peripheral blood level remained negative. Finally, baseline peritoneal fluid CEA mRNA expression was never detected, but one patient in each group became positive postoperatively. These results confirm that preoperative and perioperative mobilization of neoplastic cells is a frequent occurrence in patients with colorectal cancer, but the surgical approach (LAC vs. OC) does not seem to be a determining factor.