Ten-year trends in heart valve replacement operations

Background. There has been increasing concern in recent years about the qua lity and cost of heart valvular replacement procedures. The purpose of this study is to examine the profile of patients undergoing valvular operations during the past decade, and to look at trends in outcome and resource util ization over that period. Methods. Clinical and procedural data of 2,972 patients undergoing heart va lve replacement at Emery University Hospitals between 1988 and 1997 were re corded prospectively on standardized forms by trained medical personnel and entered into a computerized database. Results. There were 1,802 patients undergoing aortic valve replacement (AVR ), 966 undergoing mitral valve replacement (MVR), and 204 undergoing combin ed aortic and mitral valve procedures (AVR + MVR). No patients were exclude d. There was a statistically significant trend for patients undergoing AVR, MVR, or AVR + MVR over time to be older and sicker by multiple criteria. N onetheless, procedural outcome and in-hospital mortality for patients under going AVR remained unchanged. Cost and length of stay increased from 1988 t o 1992 when a concerted effort to decrease resource utilization began. Betw een 1992 and 1997 for AVR, length of stay decreased from 13.4 to 8.0 days a nd cost from $37,047 to $21,856. Similarly, between 1992 and 1997 for MVR, length of stay decreased from 15.6 to 8.1 days and cost from $45,072 to $21 ,747. The net result over the time period from 1988 to 1997 was an average decline in the cost of operation of $785 a year, adjusted for other factors . Conclusions. This study reveals that outcome of valvular replacement during the period from 1988 to 1997 has remained constant despite the patients be coming older and sicker during the same period. This constant outcome has b een accomplished, but length of stay has decreased significantly. Hospital costs increased during the first years of the study period, but then decrea sed to levels in 1997 that were equal to or significantly less than 1988 le vels. (C) 2000 by The Society of Thoracic Surgeons.