Aquatic oligochaetes are important bioindicators of sediment quality in watercourses and lakes, but their morphological identification to the species level is challenging and sometimes impossible. The use of DNA barcoding and metabarcoding could greatly facilitate the identifications of specimens and improve ecological diagnoses based on oligochaete communities. The aim of this study was to test how well metabarcoding approaches based on high throughout sequencing (HTS) could perform in assessing oligochaete species diversity and the biological quality of sediments. We analysed oligochaete communities at several sites in Swiss rivers, comparing morphological data to metabarcoding of mixed specimen, sieved sediment and total sediment samples. We amplified the cytochrome c oxidase (COI) marker using universal primers and/or specific primers to metazoans. Our results showed that metabarcoding analysis of mixed specimens allowed a more representative assessment of the biodiversity than the metabarcoding of sieved or total sediments. Although community structures obtained with the morphological and metabarcoding analyses were different in term of presence/absence of species and species abundances, the ecological diagnoses based on these two approaches largely agreed.