Abstract
BACKGROUND: The statutory health insurance system embodies a large amount of data on the treatments of their members. Depending on joint, prosthesis type, patient activity and comorbidity, knee and hip replacements can last up to 20 years. Based on statutory health insurance data the main object of this analysis was to investigate how high the early revision rate of replacements actually is. METHODS: The number of replacements in the years 2005 and 2006 has been extracted from the TK database for hip (OPS-Code 5-820, n = 20,875), knee (OPS 5-822, n = 13,466), upper limbs (OPS 5-824, n = 901), and lower limbs (OPS 5-826) replacements. This data has then been related to each consecutive operation (i. e. change or excision of joint endoprosthesis) over a joint-specific observation period of two years. RESULTS: In 3.7% of the cases joint replacements stood for less than 2 years (hip 3.5%, knee 3.8%, upper limbs 6.5%, and lower limbs 5.5%). There is a significantly positive correlation between the treatment data of the hospitals and the outcome as to low rates of reoperations at early stages. The main reason for short lifetime (76 - 81%) is mechanical failure. CONCLUSION: The percentage of joint endoprostheses with significantly short lifetimes has been unexpectedly high. The de facto lifetimes of joint endoprostheses thus often do not match the manufacturers' information. The authors strongly support the idea of a national endoprosthesis register as such a register could give detailed information on firstly whether these deficits are due to material defects, osteolysis or dislocation and secondly which products are mainly affected.