We analyzed bone turnover markers (osteocalcin, bone ALP, beta-crosslaps-CTX) and cytokines (IL-1, IL-8 and IL-10) in hip joint fluid in 10 patients before revision surgery and in 39 with idiopathic coxarthrosis. Patients with loose implants had lower concentrations of resorption marker than those with arthrosis (0.8 vs 1.3 ng/mL), but bone formation marker osteocalcin was reduced (4.2 vs 22.6 ng/mL). IL-8 and IL-10 levels were elevated in patients with implant failure (870 vs 340 pg/mL; 14.3 vs 4.0 pg/mL). We found a negative correlation between the bone resorption marker (CTX) and IL-10 in cases with prosthesis loosening and a positive correlation between IL-10 and time-to-revision. We conclude that enhanced local production of inflammatory cytokines leading to suppressed bone formation is a part of the loosening process. The expression of anti-inflammatory mediators is not sufficient to counteract the imbalance in bone turnover.