Previous hydrological trends and shifts studies carried out in Brazil used a set of stream flow records that did not covered the important critical hydrological period for Brazil-ian hydropower production occurred between 2015 and 2022. This extreme dry period has raised many substantial concerns about if it could be attributed to cyclical climate variabilities or to human induced variations. In this paper, we provide an update of trends or shifts detection studies in long-term mean annual flow (LTMAF) using a set of 52 naturalized historical time series of monthly inflows until 2022 for Brazilian hydropower plants. Another novelty of this work is the attempt to avoid errors in approximating hydrographic reconstructions of natural inflows, considering time series of inflows from hydroelectric plants located in areas with unregulated headwaters. This paper also stands out from the others by avoiding inflow data obtained using time series extensions techniques of past periods with scarcity of fluviometric records in the hydrographic region. Detection tests for assessing statistical significance of linear trends and of gradual/abrupt shifts were applied for each of the 52 selected time series. Trends and/or shifts were detected in 37 series. The results were compared with other studies that used all naturalized time series from the ONS archive.