*3.2. Meta-Analysis*

Blood volume loss in the TXA and non-TXA group varied and amounted to 0.66 ± 0.52 vs. 0.834 ± 0.592 L (MD = −0.15; 95%CI: −0.23 to −0.07; I2 = 84%; *p* < 0.001; Figure 2).


**Figure 2.** Forest plot of blood volume loss among TXA and non-TXA groups. The centre of each square represents the weighted mean differences for individual trials, and the corresponding horizontal line stands for a 95% confidence interval. The diamonds represent pooled results. Legend: CI = confidence interval; SD = standard deviation.

Drain output was reported in six studies. Polled analysis of drain output was 110.5 ± 100.4 mL for TXA group, and 222.9 ± 187.2 mL for non-TXA group (MD = −92.51; 95%CI: −141.09 to −43.93; I2 = 92%; *p* < 0.001).

Change of hemoglobin levels form preoperatively to postoperatively periods were reported in six trials and were 2.2 ± 1.0 for TXA group compared to 2.7 ± 1.1 for non-TXA group (MD = −0.51; 95%CI: −0.57 to −0.44; I2 = 0%; *p* < 0.001).

Hematocrit change was reported in five studies and was statistically smaller in TXA group (6.1 ± 2.7%), compared to non-TXA group (7.9 ± 3.1%); MD= −1.43; 95%CI: −2.27 to −0.59; I2 = 95%; *p* < 0.001).

Operation time was reported in six trials and was 89.5 ± 33.0 min for TXA group compared to 88.5 ± 32.2 min for non-TXA group (MD = −2.25; 95%CI: −4.54 to 0.05; I2 = 0%; *p* = 0.06; Figure 3). Length of hospital stay in TXA and non-TXA (control) groups was 2.1 ± 1.7 vs. 2.2 ± 1.6 days, respectively (MD = −0.15; 95%CI: −0.32 to 0.01; I2 = 0%; *p* = 0.07).

**Figure 3.** Forest plot of (**2.11.1**) operation time; (**2.11.2**) length of hospital stay among TXA and non-TXA groups. The centre of each square represents the weighted mean differences for individual trials, and the corresponding horizontal line stands for a 95% confidence interval. The diamonds represent pooled results.

Polled analysis of three studies showed that 3.0% of patients in the TXA group and 3.5% in the non-TXA group required transfusion (RR = 0.56; 95%CI: 0.20 to 1.59; *p* = 0.28). Revision was no required in TXA compared to 0.9% of cases in non-TXA group (RR = 0.33; 95%CI: 0.01 to 7.99; *p* = 0.50). Hematoma was observed in 20.4% in the TXA group and 53.8% in non-TXA group (RR = 0.39; 95%CI: 0.27 to 0.57; *p* < 0.001). Thromboembolic complications were not noted in any of the groups.
