Ecologically vulnerable areas (EVAs) are regions with ecosystems that are fragile and vulnerable to degradation under external disturbances, e.g., environmental changes and human activities. A comprehensive understanding of the climate change characteristics of EVAs in China is of great guiding significance for ecological protection and economic development. The ecosystem carbon use efficiency (CUEe) can be defined as the ratio of the net ecosystem productivity (NEP) to gross primary productivity (GPP), one of the most important ecological indicators of ecosystems, representing the capacity for carbon transfer from the atmosphere to a potential ecosystem carbon sink. Understanding the variation in the CUEe and its controlling factors is paramount for regional carbon budget evaluation. Although many CUEe studies have been performed, the spatial variation characteristics and influencing factors of the CUEe are still unclear, especially in EVAs in China. In this study, we synthesized 55 field measurements (3 forestland sites, 37 grassland sites, 6 cropland sites, 9 wetland sites) of the CUEe to examine its variation and influencing factors in EVAs in China. The results showed that the CUEe in EVAs in China ranged from -0.39 to 0.67 with a mean value of 0.20. There were no significant differences in the CUEe among different vegetation types, but there were significant differences in CUEe among the different EVAs (agro-pastoral ecotones < Tibetan Plateau < arid and semiarid areas < Loess Plateau). The CUEe first decreased and then increased with increasing mean annual temperature (MAT), soil pH and soil organic carbon (SOC) and decreased with increasing mean annual precipitation (MAP). The most important factors affecting the CUEe were biotic factors (NEP, GPP, and leaf area index (LAI)). Biotic factors directly affected the CUEe, while climate (MAT and MAP) and soil factors (soil pH and SOC) exerted indirect effects. The results illustrated the comprehensive effect of environmental factors and ecosystem attributes on CUEe variation, which is of great value for the evaluation of regional ecosystem functions.