No abstract

A few decades ago, antibiotic prophylaxis for patients with acute variceal bleeding was reported beneficial. However, endoscopic and systemic therapy for variceal bleeding has dramatically improved since then, so the necessity of prophylactic antibiotics can be questioned. In this study, we reevaluated the efficacy of antibiotic prophylaxis in acute variceal bleeding, using the most recent data in our hospital. We retrospectively analyzed the medical records of 150 patients with acute variceal bleeding who were admitted to Kurashiki Central Hospital between January 2012 and December 2016. We compared the rates of bacterial infection, in-hospital mortality, 5-day rebleeding rate, and 30-day emergency readmission between patients treated or not treated with antibiotic prophylaxis. Forty-six patients (30.7%) received antibiotic prophylaxis; 104 (69.3%) did not. The rates of the outcomes in patients with antibiotic prophylaxis were 6.5% (bacterial infection), 4.3% (in-hospital mortality), 2.2% (5-day rebleeding), and 10.9% (30-day emergency readmission) and were not significantly different form the corresponding figures in those without antibiotic prophylaxis (1.9%, 7.7%, 1.9%, and 10.6%, respectively). Moreover, these rates in our patients, even without antibiotic prophylaxis, were much lower than rates reported in past years, perhaps because of improvements in care of patients with variceal hemorrhage. Antibiotic prophylaxis was not associated with significantly better outcomes of bacterial infection, mortality, rebleeding or readmission rate in patients with acute variceal bleeding. Universal antibiotic prophylaxis for patients with acute variceal bleeding should be reconsidered.

Despite the recommendations of the latest guidelines, the practical efficacy of universal screening for identifying Lynch syndrome (LS) among patients with colorectal cancer (CRC) may be limited in the real world due to infrequent referrals and the difficulties of genetic testing. Thus, the present study aimed to retrospectively analyze the results of universal screening of patients with CRC at a referral hospital in Japan. Immunohistochemistry was performed for mismatch repair proteins [including DNA mismatch repair protein MSH6 (MSH6), mismatch repair endonuclease PMS2 (PMS2), DNA mismatch repair protein Msh2 (MSH2) and DNA mismatch repair protein Mlh1 (MLH1)] and BRAF V600E mutation. Tumors that showed the following were considered to indicate LS and patients with such tumors were designated as genetic testing candidates (GTCs): i) Loss of MSH6/MSH2; ii) loss of MSH6 alone; iii) loss of PMS2 alone; and iv) loss of PMS2/MLH1 with negative BRAF V600E. MLH1 methylation and BRAF V600E mutation were analyzed in deficient mismatch repair (dMMR) tumors retrospectively. The frequency of dMMR and GTCs in an independent cohort of patients with young-onset CRC were also investigated. Universal screening revealed dMMR tumors, GTCs and LS probands in 7.3, 3.9 and 0.4%, respectively, of 463 patients with CRC. Although dMMR tumors were observed in both younger (<50 years) and older (≥60 years) patients, the GTCs were enriched in younger individuals. Evaluation of mismatch repair status in an independent cohort confirmed the high rate of GTCs in patients with young-onset CRC. The low detection rate of LS demonstrated in this study questions the implementation of routine universal screening in regions with low prevalence of LS. Considering the enrichment of GTCs in young-onset CRCs, age-restricted strategies may be simple and efficient practical alternatives to universal screening in the real world.

No abstract