Low Risk of Variceal Bleeding Among Subjects With Cirrhosis Treated With Electroconvulsive Therapy: A Nationwide Study.

BACKGROUND AND AIM: Esophageal variceal bleeding is a dangerous complication of end-stage liver disease. There is limited information evaluating the hypothesis that medical procedures, specifically electroconvulsive therapy (ECT), may lead to variceal bleeding. The current study aims to determine the risk of variceal bleeding among subjects with cirrhosis who undergo ECT compared with other short medical procedures. METHODS: The Nationwide Inpatient Sample (2002-2013) and Nationwide Readmissions Database (2010-2014) were queried using International Classification of Disease, Ninth Revision, codes to evaluate all patients 18 years or older with cirrhosis who underwent ECT, bronchoscopy, or cystoscopy, or who experienced in-hospital seizures. Rates of variceal bleeding and hospital outcomes were compared. Multivariable analysis for readmission rate was performed. RESULTS: From the Nationwide Inpatient Sample, a total of 5,442,306 patients with cirrhosis were studied, including 840 (0.02%) patients who underwent ECT. Patients who underwent ECT were more likely to have compensated cirrhosis (P < 0.001). Among patients without ECT, 6.8% had variceal bleeding during admission compared with 0% who underwent ECT. From the Nationwide Readmissions Database, 1,383,853 patients were included, including 357 patients (0.03%) who underwent ECT during index admission. Electroconvulsive therapy did not increase the risk of 30- or 90-day readmission for variceal bleeding or mortality compared with other short medical procedures. CONCLUSIONS: Electroconvulsive therapy does not increase the risk of variceal bleeding in subjects with compensated and decompensated cirrhosis. Preoperative optimization of these patients should take the risk of bleeding into account based on current guidelines for variceal surveillance.

Base excision repair proteins are required for integrin-mediated suppression of bleomycin-induced DNA breakage in murine lung endothelial cells.

Engagement of integrin cell adhesion receptors suppresses bleomycin (BLM)-induced DNA strand breakage in endothelial cells. Previous investigation of cells from poly(ADP-ribose) polymerase (PARP)-1 knockout mice and with an inhibitor of the enzyme indicated that this facilitator of base excision repair (BER) is required for integrin-mediated suppression of DNA strand breakage. Here, small inhibitory RNA (siRNA) was used to assess the requirement for the BER proteins, DNA ligase III (Lig3) alpha, PARP-1, and X-ray repair complementing defective repair in Chinese hamster cells 1 (XRCC1), and for the long-patch BER ligase, DNA ligase I (Lig1), in integrin-mediated protection from BLM-induced DNA breakage. Murine lung endothelial cells (MLECs) were transfected with siRNA, treated with anti-beta1 integrin antibody, and then BLM. 3'-OH in DNA and accumulation of phosphorylated histone H2AX (gammaH2AX), which reflects double-strand breakage, were measured. Integrin antibody inhibited the increases in 3'-OH caused by BLM in MLECs transfected with either control or Lig1 siRNA. However, after knockdown of Lig3alpha, PARP-1, or XRCC1, suppression of DNA breakage by integrin antibody was limited. BLM increased gammaH2AX levels, and integrin treatment inhibited this by 57 to 73% in MLECs transfected with control siRNA. Integrin engagement also inhibited increases in gammaH2AX in BLM-treated cells transfected with Lig1 siRNA. In contrast, Lig3alpha, PARP-1, and XRCC1 siRNAs prevented integrin-mediated inhibition of BLM-induced gammaH2AX levels. The results suggest that the BER proteins, Lig3alpha, PARP-1, and XRCC1, are required for integrin-mediated suppression of BLM-induced DNA breakage.