BackgroundSurgeon at primary total hip arthroplasty, sometimes cannot achieve sufficient cementless acetabular press-fit fixation and must resort to other fixation methods. Despite a predominant use of cementless cups, this issue is not fully clarified, therefore we performed a large retrospective study to: (1) identify risk factors related to patient or implant or surgeon for unsuccessful intraoperative press-fit, (2) check for correlation between surgeons’ volume of operated cases and the press-fit success rate. HypothesisUnsuccessful intra-operative press-fit more often occurs in older female patients, particular implants, due to learning curve and low-volume surgeons. Materials and methodsRetrospective observational cohort of prospectively collected intra-operative data (2009–2016) included all primary total hip arthroplasty patients with implant brands that offered acetabular press-fit fixation only. Press-fit was considered successful if acetabulum was of the same implant brand as the femoral component without additional screws or cement. Logistic regression models for unsuccessful acetabular press-fit, included patients’ gender/age/operated side, implant, surgeon approach (posterior n=1206, direct-lateral n=871) and surgery date (i.e. learning curve). ResultsIn 2077 patients (mean 65.5years, 1,093 females, 1,163 right hips) were implanted three different implant brands (973 ABG-II™-Stryker, 646 EcoFit™Implantcast, 458 Procotyl™L-Wright) by eight surgeons. Their unsuccessful press-fit fixation rates ranged from 3.5 % to 23.7 %. Older age (odds ratio 1.01 [95 % CI 0.99–1.02]), female gender (2.87 [95 % CI 2.11–3.91]), right side (1.44 [95 % CI 1.08–1.92]), surgery date (0.90 [95 % CI 1.08–1.92]) and particular implants were significant risk factors only in three surgeons with less successful surgical technique (higher rates of unsuccessful press-fit with Procotyl™-L and EcoFit™ [P=0.01]). Direct-lateral hip approach had a lower rate of unsuccessful press-fit than posterior hip approach (P<0.01) but there was no correlation between surgeons’ volume and rate of successful press-fit (Spearman's rho=0.10, P=0.82). Subcohort of 961 patients with 5–7years, follow-up indicated higher early/late cup revision rates with unsuccessful press-fit. DiscussionSuccess of press-fit fixation depends entirely on the surgeon and surgical approach. With proper operative technique, the unsuccessful press-fit fixation rate should be below 5 % and the impact of patients’ characteristics or implants on press-fit fixation is then insignificant. Findings of huge variability in operative technique between surgeons of the presented study emphasize the need for surgeon-specific data stratification in arthroplasty studies and indicate the possibility of false attribution of clinically observed phenomena to patient-related factors in pooled data of large centres or hip arthroplasty registers. Level of evidenceLevel III, retrospective observational case control study.