Thirty-day mortality after total knee arthroplasty.

BACKGROUND: There have been sporadic reports on perioperative mortality associated with total knee arthroplasty. The purpose of this study was to determine risk factors for such mortality.
METHODS: A computer-assisted review of the records of 22,540 consecutive patients who had undergone total knee arthroplasty between 1969 and 1997 was performed to identify all patients who had died within thirty days after the procedure. A detailed analysis of the medical, surgical, anesthetic, and pathological records of the patients was performed, and the mortality was determined according to age, gender, diagnosis, and fixation method.
RESULTS: The rate of mortality within thirty days after the operation was 0.21% (forty-seven of 22,540). All deaths occurred in the group of 18,810 patients who had received a cemented implant, and no deaths occurred among the 3730 patients who had received an uncemented implant (p < 0.0001). The mortality rate was 0.24% (forty-three of 18,165) after primary arthroplasty and 0.09% (four of 4375) after revision arthroplasty (p < 0.0003). Three patients (0.01%) died during the operation. Forty-three of the forty-seven patients who died had a history of preexisting cardiovascular and/or pulmonary disease. Simultaneous bilateral total knee arthroplasty was associated with a significantly higher rate of perioperative mortality (p < 0.002).
CONCLUSIONS: Factors that were associated with a significantly increased mortality after total knee arthroplasty included an age of more than seventy years, primary (as compared with revision) knee surgery, use of a cemented prosthesis, preexisting cardiopulmonary disease, and simultaneous bilateral arthroplasty.