BACKGROUND: The use of an anatomic cementless stem in hemiarthroplasties for femoral intracapsular proximal fracture has been debated, notably because of bone weakness and/or morphological defects related to osteoporosis. We therefore conducted a retrospective study in subjects over 75 years of age who had received an anatomic stem partially coated with hydroxyapatite. The objectives were to determine: 1) the incidence of periprosthetic fractures (PPFs) and, 2) the influence of anatomic factors, including the Cortical Bone Ratio (CBR) (the relation between the endosteal and external diameter of the femoral diaphysis 10cm below the lesser trochanter). HYPOTHESIS: The risk of PPF with an anatomic cementless implant is greater than with cemented stems. MATERIAL AND METHODS: We retrospectively analyzed 233 patients followed up for 5 years after their surgery. The stem used was an anatomic stem with a modular neck partially coated with hydroxyapatite. The risk factors examined were age, gender, history of osteoporotic fractures, diverse causes of secondary osteoporosis, and proximal bone stock according to various referenced radiological indices such as the CBR. RESULTS: Twenty patients (15%) were lost to follow-up, 74 had died (32%) but did not undergo revision for PPF, 15 of the 139 survivors at the last follow-up (10.8%) had had a PPF, five (3.6%; four females, one male) were early fractures (≤2 months after implantation), ten (7.2%; two females, eight males) were late fractures (>2 months). Male gender was protective for PPF occurrence (RR=0.129; 95%CI (0.04-0.39); P=0.0003), whereas secondary factors of osteoporosis (RR=2.035; 95%CI (1.11-3.72); P=0.0211), and CBR>0.49 (RR=227.42; 95%CI (1.072-48,226.76); P=0.0471) were found as risk factors of PPF. DISCUSSION: The PPF rate was greater than that related to cemented stems, requiring that morphological and clinical factors of bone weakness (collected with the patient history and related to osteoporosis) be taken into account. A CBR>0.49 requires caution on the use of this type of stem. LEVEL OF EVIDENCE: Level 4. Retrospective study.