Patients in the Norwegian Arthroplasty Register with a total hip replacement (THR) have a lower long-term mortality than the age- and gender-matched Norwegian population. We analyzed the early postoperative mortality after 67,548 THR operations in 68 hospitals reported to the Norwegian Arthroplasty Register between 1987 and 1999. Data on deaths and causes of death were obtained from from Statistics Norway, and on thromboprophylaxis from a separate questionnaire sent to all hospitals. During the years 1987-2000 the 68 hospitals reported use of 6 thromboprophylaxis drugs and 24 different combinations of drugs and stockings. In 1988, only 3 of 29 hospitals reported use of low molecular weight heparin (LMWH), but in 1999, 67 of the 68 hospitals used LMWH. In the first postoperative week, the daily mortality was about 2.5 deaths per 10,000 THR patients. By the 70th postoperative day, the daily mortality had declined to about 0.57 deaths per 10,000 patients. The daily mortality of the age- and gender-matched Norwegian population was 0.95 deaths per 10,000 individuals. Early postoperative mortality increased with age, was higher in men than women, and was usually due to vascular disease. We found only a slight reduction in the 60-day postoperative mortality during the period 1987-1999. All underlying diagnoses for a prosthesis operation had a higher 60-day postoperative mortality than primary osteoarthrosis.