Objective:

2

To evaluate the effects of Aidi Injection (艾迪注射液

AD) in combination with Western medical therapies (WMT) in patients with primary liver cancer (PLC).

Methods:

2

Randomized controlled trials (RCTs) comparing AD plus WMT with WMT alone were retrieved from inception to March 2013 by retrieving the literature database thoroughly and systematically. The extracted data from included studies were analyzed and synthesized by Review Manager 5.2 software. The Cochrane risk of bias tool was used to assess the quality of included studies

and Begg's and Egger's tests were used to evaluate the potential presence of publication bias. The studies were divided into 7 separate subgroups in terms of quality of life (QOL)

recent chemotherapy and the incidence of leukocyte reduction. The subgroup analysis was applied to assess the heterogeneity between included researches

and the sensitivity analysis was used to weigh the stability of studies.

Results:

2

Twenty-four RCTs were included in this study. Compared with WMT used alone

AD as additional intervention was more effective on improving QOL (

P

<

0.01)

increasing short-term efficacy (

P

<

0.01)

prolonging life (

P

<

0.05 or

P

<

0.01)

relieving clinical symptoms (

P

<

0.01)

and reducing adverse events (e.g. reduce white blood cell counts

P

=0.002; reduce in platelet counts

P

<

0.01). Subgroup analysis showed that the hepatic artery interventions with AD was superior in improving QOL (

P

<

0.01) and enhancing short-term response rates (

P

=0.007) and reducing white blood cell counts (

P

=0.0004) than hepatic artery interventions alone (

P

<

0.01). The chemoembolization plus AD or the chemotherapy plus AD were both better than chemoembolization or the chemotherapy alone in improving the QOL and short-term response rate (

P

<

0.05 or

P

<

0.01).

Conclusions:

2

AD in combination with WMT improves QOL in patients with PLC. Considering the inherent limitations of the included studies

further well-designed

rigorously performed

high-quality

and double-blinded RCTs with large sample sizes are needed.