Study characteristics
The general characteristics of the 7 qualified articles (9 dataset) are illustrated in Table 1. Selected eligible trials enrolled 1786 participants with age ranging from 18 to 65 years old and were conducted on both genders, except for one trial that exclusively included women [25]. A number of subjects in intervention and control groups were 896 and 890, respectively. Participants were patients with overweight and obesity[20, 22-26], untreated type 2 diabetes and untreated hypertension[20], type 2 DM treated with oral antidiabetic agents (metformin alone or in combination with sulfonylureas)[21, 22], metabolic syndrome[21-23], hypertension treated with drug[22, 24], coronary heart disease concomitant cardiac dysfunction [22], hypercholesterolemia [20, 24] , PCOS [25], NAFLD [26]. All selected studies used a parallel design including a control group. Duration of orlistat therapy varied from 3 to 6 months. Daily prescribed dosage of orlistat varied between 120mg three times a day (tid) [20, 22-26] and 360mg/dl [21]. Items that participants received with orlistat during the study include: low calorie diet[20, 21], fat restricted diet and physical activity[22], low calorie/low fat diet and other medication[23-25] and in one study they didn’t take anything with orlistat[26]. The quality assessment results of the articles are shown in Fig 2.
Effect of orlistat on serum uric acid levels
7 qualified studies (9 dataset) including a total of 1786 subjects, reported serum uric acid as their outcome. Combining their findings based on random-effects model, we found that serum uric acid was significantly reduced in orlistat group compared to the control (Difference in means: -17.661μmol, 95% CI: -31.615 to -3.707, P=0.01) (Fig3), including no significant heterogeneity between studies (\(I^{2}=25.119\%\), Q-value=10.68) .We removed one study and repeated the analysis after excluding the study while the findings did not change. Based on subgroup analysis includes intervention period (3month and 6 month), sample size (< 100 and ≥ 100 participants), health status (metabolic syndrome, hypercholesterolemia, PCOS and NAFLD) were responsible, at least to some extent, for between-study heterogeneity.