BACKGROUND AND PURPOSE: Few studies have addressed the association between early migration of femoral stems and late aseptic revision in total hip arthroplasty. We performed a meta-regression analysis on 2 parallel systematic reviews and meta-analyses to determine the association between early migration and late aseptic revision of femoral stems. PATIENTS AND METHODS: Of the 2 reviews, one covered early migration data obtained from radiostereometric analysis (RSA) studies and the other covered long-term aseptic revision rates obtained from survival studies with endpoint revision for aseptic loosening. Stems were stratified according to the design concept: cemented shape-closed, cemented force-closed, and uncemented. A weighted regression model was used to assess the association between early migration and late aseptic revision, and to correct for confounders. Thresholds for acceptable and unacceptable migration were determined in accordance with the national joint registries (≤ 5% revision at 10 years) and the NICE criteria (≤ 10% revision at 10 years). RESULTS: 24 studies (731 stems) were included in the RSA review and 56 studies (20,599 stems) were included in the survival analysis review. Combining both reviews for the 3 design concepts showed that for every 0.1-mm increase in 2-year subsidence, as measured with RSA, there was a 4% increase in revision rate for the shape-closed stem designs. This association remained after correction for age, sex, diagnosis, hospital type, continent, and study quality. The threshold for acceptable migration of shape-closed designs was defined at 0.15 mm; stems subsiding less than 0.15 mm in 2 years had revision rates of less than 5% at 10 years, while stems exceeding 0.15 mm subsidence had revision rates of more than 5%. INTERPRETATION: There was a clinically relevant association between early subsidence of shape-closed femoral stems and late revision for aseptic loosening. This association can be used to assess the safety of shape-closed stem designs. The published research is not sufficient to allow us to make any conclusions regarding such an association for the force-closed and uncemented stems.