There is no consensus regarding systematic screening for infection in aseptic revision total hip arthroplasty (THA). The rationale for systematic intra-operative samples is to increase the sensitivity of latent infections detection, which may require specific treatment. However, the incidence of occult infection in revision THAs is not precisely known. As such, the aim of the study was to evaluate the incidence of occult infection in presumed aseptic revision THAs and identify associated risk factors. Bacteriological samples from 523 aseptic THA revisions performed for five years were analyzed. Revisions performed for periprosthetic joint infection (PJI) or clinical suspicion of PJI were excluded. Microbiological cultures were performed using tissue samples in 505 cases (97%), synovial fluid in blood culture vials in 158 (30%), and in dry tubes in 263 (50%). Implants were sent for sonication in 12 cases (2.6%). The mean number of microbiological samples per patient was 3.6 (range, 1-15). Histology samples were collected for 300 patients (57%). The incidence of occult infection was 7% (36 cases) and contamination 8% (42 cases). Among occult infections, the primary reasons for revision were dislocation (42%), aseptic loosening (25%), fracture (19%), and others (14%). The infection rate in the dislocation group was significantly higher than that of other reasons for revision (p < 0.001). Among the patients determined with PJI, the revision THA was performed less than a year after primary in 19 (53%). The incidence of occult PJI justifies systematic intraoperative sampling. A short time between primary arthroplasty and revision or an early postoperative dislocation is a factor to suspect infections.