The comprehensive trophic level index (TLI) is widely used in eutrophication assessment of Chinese lakes. For each sub-indices of TLI, the chlorophyll-a is the final indication due to its direct manifestation of eutrophication-related risk, and the physiochemical indicators (total nitrogen, total phosphorus, Seccchi depth and permanganate index) are indirect indications. Thus, the significant difference between the values of TLI based on the physicochemical indicators and chlorophyll-a indicates that the results based physicochemical indicators over-or underestimate the actual eutrophication status and associated risk. Here, the TLI assessment results of lakes along the middle and lower reaches of the Yangtze River based on the four physicochemical indicators and on chlorophyll-a were compared. Our results showed that, for Yangtze-isolated shallow lakes, the assessment results of TLI based on total nitrogen, total phosphorus and permanganate index underestimated the actual eutrophication status and associated risk. For Yangtze-connected shallow lakes, the results of TLI based on total nitrogen, total phosphorus, and Secchi depth overestimated the eutrophication status and associated risk, while the results based on the permanganate index underestimated the eutrophication level and associated risk. For the deep reservoirs, the evaluation of TLI based on the total nitrogen overestimated the level of eutrophication and associated risk, while the results based on the rest three physiochemical indicators underestimated the eutrophication status. The reasons for the mismatch of assessment results between the physicochemical indicators and chlorophyll-a are ascribed to two aspects. For specific lake type, some physicochemical parameters have limited indication for eutrophication risk (chlorophyll-a). Such as total nitrogen, total phosphorus, Secchi depth and permanganate index in Yangtze-connected and the total nitrogen in deep reservoirs. Moreover, the original relationships between physicochemical indicators and chlorophyll-a is mismatched. For example, the response of TP to chlorophyll-a in deep lakes is more sensitive than the counterpart used in development of TLI (TP) formula. The following suggestions were proposed for the lake eutrophication assessment of the middle and lower reaches of the Yangtze River. 1) for the lakes with adequate historical dataset, establishing the lake-specific relationships between chlorophyll-a and the four physiochemical indicators through quantile regression model and developing the lake-specific evaluate formula for physiochemical indicators; 2) Making the lake subdivision based on the water retention time and water area/water depth and deriving the type-specific relationships between chlorophyll-a and physiochemical indicators, obtaining the type-specific evaluate formula for physiochemical indicators of lakes in the middle and lower reaches of the Yangtze river. 3) In the results of eutrophication assessment, the state indicators (e.g., total nitrogen and total phosphorus) and risk indicator (i.e., chlorophyll-a) may should be evaluated separately, and more attention should be paid into the ratio of chlorophyll-a to nutrient in the lake eutrophication assessment. Meanwhile, the final indication of biological indicators needs to be emphasized. Given the fact that a general eutrophication management framework was currently implemented for Chinese lakes, which may be difficult to meet cost-efficient management objective. Thus, the type-specific and lake-specific eutrophication assessment methods proposed and reviewed in this study may be of great significance in future lake management.