The liver is a central organ in metabolism and there are numerous inherited metabolic disorders that have their origin in this organ. Gene therapy represents a promising therapeutic approach for this type of diseases. Although in preclinical studies using AAV-based vectors high and long-term hepatic expression has been achieved, there are still some concerns about the expression levels and the percentage of hepatocytes that can be transduced. This is of particular importance for diseases in which the deficiency of an enzymatic activity is associated with the generation of toxic products that require the transduction of a high percentage of hepatocytes. Therefore, the main goal of our study was to improve liver transduction efficacy by an AAV serotype. For this purpose, the vector was administered directly to the liver via suprahepatic veins (SHV) using a catheter and balloon occlusion or via the hepatic artery (HA) with balloon occlusion of the suprahepatic vein. The experiment was performed in Macaca fascicularis, with 8 animals being infected with a dose of 3×1013 gc/kg of an AAV5 expressing the reporter gene hSEAP (human secreted embryonic alkaline phosphatase) under the control of a liver specific promoter (AAV5-AAT-hSEAP). The first group (2 animal) received the virus systemically in the saphenous vein, the second group (3 animals) received the virus via the SHV with 10 minutes of balloon occlusion and the third group (3 animals) via the HA with 10 minutes of balloon occlusion of the SHV. The procedure could only be performed on the right branch of SHV and HA due to the small size of the left SHV and HA branches, reaching only 40% of the liver. hSEAP-specific activity in serum samples of the groups that received the virus directly in the hepatic blood flow was higher compared to the animals receiving the vector by peripheral intravenous injection. This was correlated with the presence of a higher number of viral genomes in the liver of the animals. Moreover, when their distribution was analyzed we found that the animals that had received the virus through the SHV or HA displayed higher infection rates in the right side of the liver, where the administration had been performed. In conclusion, direct administration of the vector AAV5-AAT-hSEAP through the hepatic blood flow with balloon occlusion was associated with higher transgene expression and an increase in vector genomes in the injected area. Studies in bigger animals in order to reach both sides of the liver should be performed in order to explore the potential use of alternative routes of administration in the clinic.