2 | MATERIALS AND METHODS
2.1 | Search strategy
Study search and data collection was performed from June to September 2021. Major online medical databases such as PubMed, Embase, Scopus, Web of Science and Google Scholar were searched to gather all studies in which the variations, maximal diameter and ostial area of the PVs were investigated. The following search terms were used utilizing the Boolean technique: (pulmonary veins anatomy OR pulmonary veins anatomy variation OR accessory pulmonary veins OR anomalous pulmonary venous return OR pulmonary veins origin) AND (pulmonary veins ostial area) AND (pulmonary veins diameter OR pulmonary veins width). Neither date, language, article type nor text availability conditions were applied. An additional search through the references of the identified studies was conducted at the end of the search stage to ensure the accuracy of the process. During the study, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed. Additionally, The Critical Appraisal Tool for Anatomical Meta-analysis (CATAM) was used to provide the highest quality findings .
2.2 | Eligibility assessment
The database search and the manual search identified a total of 9119 studies which were initially evaluated by two independent reviewers. After removing the duplicates and irrelevant records, a total of 1573 articles were qualified for a full-text evaluation. To minimize the potential bias and to maintain an accurate statistical methodology, articles such as case reports, case series, conference reports, reviews, letters to the editors and studies that provided incomplete or irrelevant data were excluded. Inclusion criteria involved original studies with extractable numerical data regarding variations, maximal diameter and ostial area of the PVs. As a result, 73 studies were included for variations analysis, 86 for diameter analysis and 18 for ostial area analysis. The AQUA Tool which was specifically designed for anatomical meta-analyses was used to minimize the potential bias of included studies . The flowchart outlining the study inclusion process is shown in Figure 1.
2.3 | Data extraction
Data from studies that met inclusion criteria were extracted by two independent reviewers. Qualitative data, such as year of publication, country and continent of origin, methodology of data collection, information about the diseases in the studied groups were gathered. Quantitative data, such as sample size, numerical data regarding prevalence among variations of the PVs, mean maximal diameter of every PV with its standard deviation or interquartile range and mean ostial area of every PV with its standard deviation or interquartile range were also extracted. Studies containing mean results, but without standard deviation or interquartile range or unclear or unspecified variations were excluded. Any discrepancies between studies identified by the two reviewers were resolved by contacting the authors of the original studies wherever possible or by consensus with a third reviewer.
2.4 | Statistical analysis
To perform the meta-analyses, STATISTICA version 13.1 software (StatSoft Inc., Tulsa, OK, USA), MetaXL version 5.3 software (EpiGear International Pty Ltd, Wilston, Queensland, Australia) and Comprehensive Meta-analysis version 3.0 software (Biostat Inc., Englewood, NJ, USA) were used. A random-effects model was used in all analyses. Chi-squared test and I-squared statistic were employed to assess the heterogeneity among the studies . A p-value and confidence intervals were used to determine statistical significance between studies. A p-value lower than 0.05 was considered statistically significant. In case of overlapping confidence intervals, differences were considered statistically insignificant. I-squared statistics were interpreted as follows: values of 0-40% were considered as “might not be important”, values of 30-60% were considered as “might indicate moderate heterogeneity”, values of 50-90% were considered as “may indicate substantial heterogeneity” and values of 75-100% were considered as “may indicate substantial heterogeneity”.