Meta-anlysis Of Statins In The Treatment Of Pneumoniaor Or Sepsis

Objective：Application of meta analysis to systematically review the effective oftherapy of statins for the patients of pneumonia or sepsis.Statinpleiotropic effects can interpretation the association of pneumonia orsepsis,and to provide evidence-based basis for the using of statin forpneumonia or sepsis.Materials and methods：1. Electronic databases including PubMed,Embase,The Cochrane Library, CNKI, WanFang Data and VIP were searched from inception to January31,2015to identify relevant studies.We used a combination of keyword related to the type of statin "hydro-xymethylglutaryl coenzyme A redu-ctase inhibitors" or "anticholesteremic agents" or "statin" or "simvastatin"or "atorvastatin" or "fluvastatin" or "lovastatin" or "pitavastatin" or "pravastatin" or "rosuvastatin") and the type of infection-associated disease ("infection" or "sepsis" or "bacteremia" or "pneumonia").2. Randomized controlled trials are all enrolled in the study,selectedobject is statin therapy in hospitalized patients with pneumonia orsepsis,The primary outcome were in-hospital mortality,C reactive protein,the rates of mechanical ventilation usage or ICU admission.3. The following inclusion criteria were:①a randomized controlledtrial study;②Meet the diagnostic criteria of pneumonia orsepsis;③r esearch for clinical trials of statins;④data available on theprimary outcomes.The exclusion criteria were:①data is inco-mplete andcannot take advantage of the literature;②D uplicate publication orMultiple contributions. 4. The methodological quality of studies included was evaluated usingthe Jadad scale.5. Statistical analysis was performed using Review Manager5.3.Theinverse variance were expressed as risk ratio (RR) with pertinent95%confidence intervals(CI);Mean differences(MD)and95%confidenceintervals(CI) were expressing the continuous variables. Heterogeneityacross trials was assessed using a standard chi-squared test and I2statistic,with significance being set at P less than0.10or I2more than50%.Andusing the random effects modelfor statistical analysis. We furtherconducted subgroup analyses to explore possible explanations forheterogeneity. Publication biases were evaluated using Funnel plots andBegg′s test，Egger′s test。Otherwise, using the fixed effects model. A P-value of less than0.05was considered to represent statistical significance.Results：1. A total of6studies including984patients were included in the metaanalysis,compared with the placebo group, statins in improving mortalityof hospitalized patients with pneumonia or sepsis had no significantdifference; RR=0.83;95%confidence interval:[0.63-1.08], I2=35%, P=0.17.2. A total of7studies including1053patients were includedin the meta analysis, Compared with the placebo group, statins inimproving the rates of mechanical ventilation usage or ICU admission ofpatient infected had no significant difference; RR=0.99;95%confidenceinterval:[0.96-1.03], I2=0%, P=0.72.3. A total of4studies including426patients were included in the metaanalysis, Compared with the placebo group, statins can significantlyreduce levels of CRP; MD=-7.81mg/L;95%CI:[-10.10,-5.51], I2=52%, P <0.001. Conclusion：Our results form Meta-analysis suggest that statins can notsignificantly reduce both in-hospital mortality and the rates of mechanicalventilation usage or ICU admission,but can significantly reduce plasmaCRP levels, which can reduce inflammation improve to improve theprognosis.However,with limited researches, smaller sample study, shorterfollow-up time, and as a result,further randomized controlled trials andlarger sample are necessary to further support this conclusion.