2 Methods
2.1 Literature search
This present analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement6. Two of the authors (YCL and LJH) searched PubMed, EMBASE, and the Cochrane Library independently and extensively for articles of interest published before March 2020. The keywords used in the search process included “stapler” or “stapling” or “mechanical” and “laryngectomy”. Moreover, reference lists of the included studies were also reviewed to identify additional studies.
2.2 Study selection and data extraction
The inclusion criteria were studies including only patients with laryngeal cancer who underwent TL, articles published in English, and studies comparing the outcomes of TL between the SC and MC techniques. The exclusion criteria were based primarily on the absence of one of the inclusion criteria. Studies without a control group, studies using the same database, articles not published in English, duplicate studies, case reports, abstracts, letters to the editor, and articles pending publication of the full text were excluded from the present analysis. Data were independently extracted by the 2 researchers (YCL and LJH). The bias in the studies was assessed independently by two researchers (YCL and LJH) using the Newcastle-Ottawa Scale and the Cochrane Collaboration’s risk of bias tool (RoB 1.0) for nonrandomized and randomized studies, respectively 7,8. Discrepancies in study bias classification were discussed between the two authors until consensus was achieved.
2.3 Outcomes
The main outcomes of this study included operative time, pharyngocutaneous fistula (PCF), length of hospitalization and postoperative surgical site infection.
2.4 Data analysis
The results were analyzed using Comprehensive Meta-Analysis software (Version 3; Biostat, Englewood, NJ). Mean differences (MDs) were calculated to compare the operative time and the length of hospitalization between the SC and MC groups. Odds ratios (ORs) were calculated to compare the PCF rate and postoperative surgical site infection rate between the SC and MC groups. The overall effect was pooled using a random-effects model. Statistical heterogeneity among studies was measured using the I2 statistic, which calculated the proportion of overall variation attributable to between-study heterogeneity. An I2 statistic exceeding 50% indicated moderate heterogeneity, and anI2 statistic exceeding 75% indicated high heterogeneity 9. Potential publication bias was assessed using a funnel plot and the Egger intercept test9. A 2-sided P value < 0.05 was considered statistically significant.