RESUMO
Objective To systematically evaluate the therapeutic effect of pulse high volume hemofiltration (PHVHF) for treatment of patients with sepsis. Methods Databases such as PubMed in American National Medical Library, Holland medical abstract Embase, the Cochrane Library, China National Knowledge Internet (CNKI), China Biological Medical Literature Database (CBM), VIP, WanFang databases, etc. were searched by computer to retrieve randomized controlled trials (RCTs) on PHVHF for treatment of patients with sepsis, and the retrieval time ranged from the creation of database to March 25, 2017. Both groups of patients received conventional treatments, including antibiotics, fluid resuscitation, vasoactive agents as well as other organ function support treatments to maintain the basic vital signs stable. Patients in PHVHF group received PHVHF besides conventional treatment, while the patients in control group received conventional treatment or any other continuous renal replacement therapy (CRRT) mode with PHVHF excluded. The literatures accepted should at least include one of the following changes of outcome indicators, such as overall mortality, the levels of inflammatory mediators eliminated, acute physiology and chronic health evaluation Ⅱ(APACHE Ⅱ) score, service life of filter, amount of replacement fluids used. Two researchers independently screened literatures, extracted data, and assessed the methodological quality of included studies. Meta-analysis was conducted by using RevMan 5.3 software and the publication bias was evaluated by visually inspecting funnel plots. Results A total of 11 RCTs involving 410 patients met eligibility criteria, of which 204 patients in the PHVHF group and 206 patients in the control group. In the control group 5 RCTs used other CRRT modes, and 6 RCTs applied the conventional therapy. Meta-analyses showed that interleukin-6 [IL-6, standard mean difference (SMD) = -0.80, 95% confidence interval (95%CI) = -1.56 to -0.06, P = 0.04], tumor necrosis factor-α (TNF-α, SMD = -0.78, 95%CI = -1.33 to -0.23, P = 0.006), APACHE Ⅱ scores [mean difference (MD) = -3.80, 95%CI = -5.08 to -2.52, P < 0.000 01] were obviously lower than those in control group, but no significant statistical significance in mortality was seen between the two groups [relative risk (RR) = 0.72, 95%CI = 0.49 - 1.07, P = 0.10]. Further subgroup analyses suggested that compared with conventional treatment group, in PHVHF group mortality (RR = 0.40, 95%CI = 0.16 - 0.95, P = 0.04), IL-6 (SMD =-1.87, 95%CI = -3.58 to -0.16, P = 0.03), TNF-α (SMD = -1.32, 95%CI = -2.24 to -0.40, P = 0.005), and APACHE Ⅱscore (MD = -4.29, 95%CI = -6.02 to -2.56, P < 0.000 01) were significantly decreased; however, only a significantly decreased APACHE Ⅱ score (MD = -2.95, 95%CI = -4.56 to -1.35, P = 0.000 3) was observed in PHVHF group compared to that in subgroup of other CRRT modes. Conclusions Compared with using conventional therapy alone, using PHVHF combined with conventional therapy for treatment of patients with sepsis can more effectively improve their prognosis, and PHVHF can be the efficacious alternatives of other CRRT modes especially the HVHF. However, due to the limited quantity and quality of the included studies, further high-quality, multicenter, large-scale RCTs are needed to verify the above conclusion.
RESUMO
BACKGROUND: Despite the development of medical knowledge and technology, it has long been pointed out that the treatment guidelines are not sufficiently based on evidence. It has not been yet studied how evidence based medicine is implemented when physicians make their therapeutic decision. The purpose of this study was to determine the degree of evidence based interventions in a university based family medicine outpatient clinic. METHODS: The degree of evidence based practice was evaluated using Ellis and Gills' method developed by the Evidence - Based Medicine Center in Oxford. The Patients' records of an outpatient clinic of a university hospital were reviewed on the primary diagnosis - intervention. The evidence based guidelines were defined as traditional textbooks and the results of randomized controlled trials found on databases such as Medicine, Clinical evidence, Best evidence, and Cochrane. ln case where there were no guidelines, consultation with the specialists was done. The degree of the evidence based therapeutic interventions was assessed by three levels. RESULTS: There were 179 primary diagnosis - intervention pairs, among them, 125 pairs (69.8%) of interventions were based on randomized controlled trial evidence and 19 pairs (10.6%) based on convincing non-experimental evidence. No evidence was found for 35 pairs (19.6%) . As a result, 80.4% of the total 144 pairs were regarded as evidence - based medicine. CONCLUSIONS: The result showed that considerable portion of the total cases were evaluated as based on clinical evidence, which is similar to the conclusions of the previous studies in other countries. We hope that future similar studies will be conducted in other institutions as well as in other specialities.