The changing role of surgery in the management of infective endocarditis.

IN THE PAST THREE DECADES a striking change has occurred in the microbiological spectra, clinical features and natural history of infective endocarditis. Undoubtedly, the development of a host of potent antimicrobial agents and their widespread availability and, at times, indiscriminate use, have played a key role. Other important factors include the use of prosthetic heart valves, the increasing use of immunosuppressant agents and the epidemic of intravenous drug abuse in urban centers. It is ironic that cardiac surgery in the uninfected heart has created the environment for a significant increase in the number of cases of infections of the heart. As improvements in the design of both mechanical and tissue valves have occurred, providing enhanced long-term survival, and as the immediate operative risk has decreased, a large population of patients has been spawned who currently have one or more prosthetic valves. These valves, however, may be the source of various morbid events, among which is infective endocarditis. At one center, approximately one-third of those found at autopsy to have infective endocarditis had a prosthetic heart valve.' The irony lies in the fact that prosthetic valve replacement is increasingly being used therapeutically to manage prosthetic valve infection. The changing clinical picture of infective endocarditis is startling. The classic pattern of subacute bacterial endocarditis characterized by a protracted history of constitutional symptoms including lowgrade fever, night sweats, weight loss, general malaise, easy fatigability, arthralgia and pallor, coupled with changing cardiac murmurs and evidence of small, repeated peripheral embolizations in a patient with known pre-existing heart disease, is infrequently seen. Today, in contrast, the patient is likely to be acutely toxic and far more febrile. He tends to be ill for less time before he seeks medical help, frequently has either a prosthetic heart valve or no known previous cardiac abnormality, often is an intravenous drug abuser (particularly in an urban setting), may present without a murmur but with septic pulmonary emboli reflecting tricuspid valve involvement or a major peripheral embolus (particularly if the microorganism is a fungus), and, finally, may develop acute pulmonary edema and/or peripheral vascular collapse reflecting sudden hemodynamic deterioration. The difference in the clinical picture of infective