2.6. Statistical analysis
Statistical analyses were performed using SPSS version 23 (IBM Corp. in Armonk, NY). There was no missing data of the variables of interest for the patients included in the study. Shapiro-Wilk test was used to evaluate the distribution of the data. Descriptive data are present as the median with the interquartile range (IQR) for non-normally distributed numerical variables, and the frequency (n) and the percentage (%) for categorical variables. The Mann–Whitney U test was used for comparing non-normally distributed data between two study groups. Pearson Chi-square test was used for comparing categorical variables. The recurrence risk was determined by univariate and four multivariate logistic regression models that were developed to estimate the recurrence risk using the demographics, baseline characteristics, perioperative characteristics and presence of postoperative complication as potential confounders. In the model 1, age and sex were included; in the model 2, sinus type and number of orifices were included in addition to variables of model 1; in the model 3, duration of surgery, length of stay and wound healing time were included in addition to variables of model 2; in the model 4, presence of any type of complication was included in addition to variables of model 3. The Odds Ratios (ORs) with 95% confidence intervals (CIs) were assessed for comparing recurrence risk between surgical procedures. p <0.05 was considered as statistically significant level.