Pharmacological Prevention For Stress Ulcer Bleeding: A Systematic Review Of Randomized Controlled Trials

Background: Some randomized trials in the prevention of stress ulcer have manifested that routine prophylaxis should be administered to a large proportion of critically ill patients in intensive care units (ICU ) . However , no individual study has definitively established whether these agents (PPI , H₂RA or sucralfate ) decrease either stress ulcer bleeding or mortality , nor has any study elucidated the relative merits or risks of different prophylactic regimens .Objective: The objectives of the systematic review are to assess the efficacy and safety of pharmacological prevention for stress ulcer bleeding (SUB) in critically ill patients.Materials and Methods: Trials were identified by searching for Cochrane Controlled Trials Register , Medline , Embase and the Chinese Biological Medicine Database (CBM-disc) , Chinese VIP Database and Chinese EBM Database . We handsearched the data from the proceedings of correlated conferences , eight kinds of important Chinese journals and the references of all included trials . Two reviewers assessed the quality of studies , extracteddata independently . Disagreement were resolved by discussion or the third party if needed . The following primary outcomes were assessed : 1) Incidence of SUB ; 2) Incidence of NP ; 3)Mortality ; 4)Adverse events . Secondary outcomes were : 1) Gastric Ph ; 2) Length of hospital stay ; 3) Length of ICU stayResults: Forty-seven trials involving 5586 patients were included. Most trials were of poor quality. Meta-analysis was performed. (1) About SUB , PPI(OR 0.13, 95% CI 0.09 to 0.19; pO.00001, NNT=3^ H2RA(OR0.30, 95% CI 0.23 to 0.39; pO.00001, NNT=6) significantly reduced the incidence of SUB in comparison to placebo or no prophylaxis group . PPI significantly decreased the incidence of SUB in comparison to H2RA (OR 0.25, 95% CI 0.18 to 0.34; pO.00001). There was insignificant difference in the incidence of SUB at the end of the prevention between H2RA and sucralfate(p=0.28 ) . There was a statistically insignificant trend in favor of PPI (p=0.63) and H2RA (p=0.11) in clinically important bleeding and there was insignificant difference in the incidence of clinically important bleeding between H2RA and sucralfate (p=0.72) . (2) About NP , H2RA significantly increased the incidence of NP in comparison to sucralfate (OR 1.55, 95% CI 1.15 to 2.10; p=0.004). There was insignificant difference in the incidence of NP between PPI and H2RA (p=0.80), (3) About mortality, PPI(OR0.41, 95% CI 0.25 to 0.67; p=0.0003, NNT=11^ H2RA(OR 0.62, 95% CI 0.44 to 0.87; p =0.005, NNT=19) significantly decreased mortality of disease in comparison to placebo or no prophylaxis . There wasn't significant difference betweent H2RA and sucralfate (p=0.44) . Additionally H2RA significantly decreased mortality of SUB in comparison to placebo or no prophylaxis(OR 0.15, 95%CI 0.05 to 0.48;p=0.001,NNT=17) . However , there was a insignificant difference in mortality of SUB betweent PPI and H2RA (p=0.05). (4) About adverse events , PPI group were safer than H2RA group (OR 0.17,95% CI 0.04 to 0.77;p = 0.02) . (5) About gastric pH , We did not performe Meta-analysis of gastric pH for discrepancy in methodology. Conclusions: PPL H2RA significantly reduced the incidence of SUB and mortality of disease and insignificantly reduced the incidence of clinically important bleeding . PPI significantly decreased the incidence of SUB comparison to H2RA . But the data provided no evidence suggesting differential effectiveness of H2RA vs sucralfate in the incidence of SUB > mortality of disease and clinically important bleeding . H2RA could significantly reduce the incidence of mortality of SUB . However , there was a insignificant difference in mortality of SUB betweent PPI and H2RA . Additionally H2RA might significantly increase incidence of NP in comparison to sucralfate . PPI and H2RA administered intravenously were more effective than that administered orally or by nasogastric tube . So it was suggested that PPI and H2RA should be administered intravenously in clinical application . PPI were safeter than H2RA . But the application of conclusion must be cautious because the trails published were not good quality . More randomized controlled trials with enough sample size^ uniform standard ^ further high-quality and scientifically sound methodology should be performed.