PURPOSE OF THE STUDY Non-operative and operative treatment of acetabular fractures is associated with a risk of development of posttraumatic avascular necrosis of femoral head or with the development of posttraumatic coxarthrosis. The purpose of the study was to identify the occurrence of these two complications in patients in our group and to determine the risk factors for the development of these complications. MATERIAL AND METHODS The retrospective study was conducted in two Level I trauma centres in the period from 2009 to 2014. The group included patients with an acetabular fracture. The inclusion criteria were the following: outpatient follow-up for the period of 3 years after the injury, full radiology and CT documentation. The exclusion criteria were the following: pathological fractures, missing documentation after the union, insufficient follow-up period or dissent of the patient. The inclusion criteria were met by 192 patients, 48 women and 144 men, with the mean age of 48.9 years. The following basic epidemiological data were monitored: age, sex, cause of injury, type of fracture according to the Letournel classification, occurrence of associated injuries and type of therapy. The patients undergoing non-operative treatment as well as patients undergoing operative treatment underwent clinical and radiological examinations at 3 and 6 weeks after the injury, then at 3, 6 and 12 months after the injury, subsequent follow-up checks were done at a year-interval up to 3 years after the injury. RESULTS The posttraumatic avascular necrosis of femoral head developed in 22 patients (11.7%, 17 men, 5 women, p = 0.1159), with the mean age of 55.3 years (STDEVP 15.5, range from 22 to 82). The average time to the development of femoral head necrosis was 13.1 months (STDEVP 17.0, range from 1 to 80), median 6 months, 95% percentile 34 months. In a total of 16 patients necrosis developed within 18 months after injury, while in 6 patients after a longer period of time. Progression of coxarthrosis was observed in 63 patients (33.5%, 44 men and 19 women, p = 0.0447). Within 24 months progression was seen in 55 patients, beyond 2 years in 8 patients. Confirmed as risk factors for the development of posttraumatic avascular necrosis of femoral head and progression of posttraumatic coxarthrosis were the age 60 years and above (p = 0.0023), posttraumatic medialisation of the femoral head greater than 2 mm (p < 0.0001), displacement in the weight bearing area within the acetabulum greater than 2 mm (p < 0.0001), operative treatment (p = 0.0014), combined surgical approach (p = 0.0044), and higher caput-collum-diaphyseal (CCD) angle of proximal femur (p = 0.0142). At risk for the development of avascular necrosis were the A5 type fractures (p = 0.0214) and B2 type fractures (p = 0.0218), at risk for the development of coxarthrosis were the C1 type fractures (p = 0.0122). The isolated fractures of the anterior column were by contrast associated with a significantly lower risk for development of both the AVN (p = 0.0052) and posttraumatic coxarthrosis (p = 0.0006), the isolated fractures of the posterior wall were associated only with a higher risk for AVN and coxarthrosis summation (p = 0.0399), and the same applies to the T fractures (B3, p = 0.0200). DISCUSSION Majority of current studies regarding acetabular fractures focuses on operative treatment, short-term complications and comparison of outcomes of operative and non-operative treatment. Only a few studies are dedicated to epidemiological data, or risk factors for the development of medium-term and long-term complications. In the presented study attention was paid to two main complications arising from these fractures and requiring subsequent operative treatment: posttraumatic avascular necrosis of femoral head and posttraumatic coxarthrosis. The limitations of the study are its retrospective nature, summation of groups from two trauma centres (potential bias in patient enrolment or in assessing radiographs), lower frequency of clinical surveillance in non-operatively treated patients after healing, a fairly low number of non-operatively treated patients - especially those with osteoporosis-related insufficiency fractures. Ranking among the relative limitations is also the Letournel classification which, though most commonly used at present, shows a low level of correlation in comparisons by more evaluators. CONCLUSIONS Confirmed as significant risk factors for the development of posttraumatic avascular necrosis of the femoral head and posttraumatic coxarthrosis progression were the age of 60 and above, posttraumatic medialisation of the femoral head greater than 2 mm, displacement involving the weight bearing area of the acetabulum greater than 2 mm, operative treatment, combined operative approach. At risk are also the transverse fractures (A5 according to the Letournel classification), transverse posterior wall fractures (B2 according to the Letournel classification) and at risk for the development of coxarthrosis are both-column fractures with the high fracture line of the anterior column (C1 according to the Letournel classification). Femoral neck valgosity was a risk factor for the development of femoral head necrosis. Conversely, sex and instability of osteosynthesis detected on the radiograph within 3 months postoperatively were not confirmed as the risk factors for the development of aforementioned complications. To verify the results of this retrospective study other multicentric and prospective studies should be conducted. Key words: complications of treatment of acetabular fractures, risk factors for avascular necrosis of femoral head, risk factors for coxarthrosis.