RESULTS OF SYSTEMATIC REVIEW AND META-ANALYSIS
Based on the selection criteria and after a systematic review followed by a meta-analysis, as shown in Table 1, the literary findings included the main information from the studies to the techniques used for relapsed weight loss(RWL) after bariatric surgery, rate of success (SR), type of study, number of patients, references and follow-up. A total of 34 studies were found, of which eight (8) papers were published using the argon plasma method (APC) to reduce gastrojejunal anastomosis, twenty (20) papers using the suturing procedure (SUT), and 6 papers using techniques surgeries (SUR).
The intervals in the number of patients evaluated in each procedure were 10 to 558 for APC, 3 to 252 for SUT, and 5 to 116 for SUR. Most studies followed prospective longitudinal studies. The randomized studies were related to one study on (TORe), one on EndoCinch®, and one on APC (with crossover) (Table 3). The intervals of the RWL and SR averages of the procedures are shown in Table 3. Thus, for RWL, the averages for APC, SUT, and SUR were respectively 19.39 ± 10.94, 26.06 ± 16.71, and 17.05 ± 9.85, and for SR, the means were, respectively, 89.75 ± 16.75, 86.80 ± 11.44, and 76.00 ± 21.57.
Also, the present study presented the results of the mean, standard deviation, and minimum and maximum values ​​and p-value values ​​of the normality test, with p> 0.05 being normal (95% CI), of the APC, SUT, and SUR procedures to RWL and SR. Thus, the results in relation to the RWL showed that the APC and SUR studies showed Gaussian (normal) behavior, with p = 0.253> 0.05 and p = 0.735> 0.05, respectively. The studies on SUT showed non-normal behavior, with p <0.005, as shown in Table 4. The results about SR showed that only SUR studies showed Gaussian behavior, with p = 0.074> 0.05 (Table 5).
The equality of variance test was also performed for further Tukey analysis between the APC, Suturing and Surgery procedures to the RWL and SR of the post-bariatric surgery procedures, with p>0.05 without statistically significant difference (95% CI). Due to the results, they presented equality of variances, both for RWL, with p = 0.707> 0.05 (Figure 2), and for SR, with p = 0.174> 0.05 (Figure 3), followed by Tukey’s analysis (ANOVA). This compared the averages of the percentage of RWL and SR of the procedures, with p> 0.05 without a statistical difference (CI 95).
Thus, in relation to the RWL, Figure 4 shows that the difference between the means of the SUT vs. APC, SUR vs. APC and SUR vs. SUT were respectively 6.67 (p = 0.843), -2.34 (p = 0.878) and -9.01 (p = 0.510). Figure 5 shows for SR that the difference between the means of the SUT vs. APC, SUR vs. APC and SUR vs. SUT were respectively -2.95 (p = 0.880), -13.75 (p = 0.284) and -10.80 (p = 0.343). Therefore, the results of RWL and SR showed a high association in each comparison between the groups paired above, with R2 ≥ 50% (homogeneity).