Review manager 5

Using Endnote X8, two researchers examined the titles and abstracts of the papers and then screened them according to the inclusion and exclusion criteria. Articles that met the requirements were further evaluated by reading their full text. In a disagreement between the two researchers, a third researcher passed the final judgment.
The selected documents were thoroughly reviewed, and all their information was entered into a data extraction form; then, the data were imported into Microsoft Excel. In the next step, the data were transferred from Excel to Review Manager 5.3 and Stata 14. The data collected in this study included the author's name, year of publication, location of research, number of patients, mean age, duration of follow-up, design, female/male, deceased donors/living donors, number of UTIs, risk factors of UTI including underlying disease (diabetes, hypertension), use of ureteral stents, days of catheterization, history of UTI, acute rejection process (ACR), abnormal anatomy of the urinary tract, and the abundance of UTI-causing bacteria. The primary objective was to investigate the prevalence of UTI in kidney recipients, and the main goal was to examine the risk factors of UTI in these patients.

Forest plots were used to assess and summarize the multivariate HRs to describe the relationships between sex differences and CSS, OS, RFS, and PFS. Studies were not considered eligible for meta-analysis if they had used Kaplan–Meier log-rank, univariate Cox proportional hazard regression, or general logistic regression analyses. For all studies reporting only HRs and P values, the corresponding 95% CIs were calculated [21 (link), 22 (link)]. The studies included in the meta-analysis were evaluated for heterogeneity in outcome using the Cochrane’s Q test and the I² statistic. Significant heterogeneity was indicated by a P < 0.05 in Cochrane’s Q tests and a ratio of > 50% in I² statistics. Fixed-effects models were used to calculate pooled HRs for non-heterogeneous outcomes [23 (link)–25 (link)]. Sensitivity analyses were conducted to assess the robustness of the results based on the quality of the studies included. All statistical analyses were performed using Review Manager 5.3 and Stata/MP 14.2 (Stata Corp., College Station, TX) with the level of statistical significance set at P < 0.05.

Data were pooled by Review Manager 5.3 software (Cochrane Collaboration 2014, Copenhagen, Denmark) and analyzed with random- or fixed-effects models. Data extracted were continuous and were therefore summarized as mean differences (MDs) with 95% confidence intervals (CIs) and compared between groups. Subgroup analyses were also performed to compare the effects of different doses and treatment durations. Statistical heterogeneity among studies was assessed using Q tests; the degree of heterogeneity, using the I² value. Sensitivity analysis was conducted by excluding one study at a time and assessing whether the pooled results from the remaining studies differed from the results obtained across all studies. Publication bias was assessed using the Egger’s test in Stata software, version 13.1 (Stata Corp, College Station, TX, USA) and using a funnel plot generated by Review Manager 5.3 software. Admittedly, the funnel plot and Egger’s test is less reliable for the small number of studies in this meta-analysis [28 (link)].

The data were analyzed with the Cochrane Review Manager 5.3 and STATA 11.0 software according to the preferred reporting items for systematic reviews and meta-analysis (PRISMA) statement31 (link)32 (link). Risk ratios (RR) were calculated and pooled with a 95% confidence interval (CI) for dichotomous variables. The heterogeneity was estimated using the I² test, which was considered to be low heterogeneity when I² ≤ 25%. A fixed-effects random effects model was used if the I² was ≤25%. Otherwise, a random effects model was applied. We used the funnel plot and Eger’s test to assess potential publication bias33 (link).

The original studies included used the mean and SD to assess the MDI and PDI of the infants. We pooled the MDI and PDI scores of each study separately using the DerSimonian-Laird formula (random-effects model) [31 ]. Statistical heterogeneity [32 (link)] between the studies was assessed using the Q and I² statistics. Values of p<0.1 and I²>50% indicated high heterogeneity [2 (link)]. We conducted a subgroup analysis based on the type of chorioamnionitis (clinical or histologic) as well as a stratified analysis of the effect of maternal chorioamnionitis on the MDI of the infants based on the blinding of the medical history in the MDI assessment (blinded or not blinded), the MDI assessment after correction for gestational age (the MDI of the infants assessed at the correct age or the MDI of the infants assessed at an incorrect age), and the time of the MDI assessment (at 7 months, 12 months, or 18–24 months). We performed sensitivity analyses by omitting one study at a time.
We used a funnel plot to assess the publication bias. We used Egger’s [33 (link)] and Begg’s [34 (link)] tests to assess the publication bias, which was considered to be statistically significant when p<0.05. We performed the statistical tests using Stata software, version 12.0 (StataCorp, College Station, TX) and Review Manager 5.3.

A pooled calculation was performed on the renal parameter of eGFR or ACR. Weighted mean difference (WMD) and 95% confidence interval (CI) were calculated for changes of eGFR and ACR. A fixed- or random-effects model was used according to the heterogeneity, which was quantified by the index of I². Sensitivity test was used to examine the influence of individual study on an overall estimate. In case of possible important heterogeneity, subgroup analysis was accordingly performed on related parameters. Publication bias was also examined by Begg’s and Egger’s tests if there were at least five studies reporting changes of eGFR or ACR. All these analysis were performed by using Review Manager (5.3) and STATA (12.0) software.