OJAI, CALIF. — Surgical treatment for infective endocarditis is notoriously risky, with mortality up to 20% in some studies and morbidity much higher. But a new scoring system designed to identify the 13 most significant risk factors for morbidity and mortality might help guide clinical decision making.
Using data from the Society of Thoracic Surgeons Adult Cardiac Surgery Database, a research team led by Dr. Jeffrey G. Gaca of Duke University Medical Center, Durham, N.C., conducted a multivariate analysis to identify the risk factors and from that, they developed a simple bedside scoring system, Dr. Gaca said at the meeting.
The investigators identified every patient in the STS database who underwent surgery on the aortic, mitral, or tricuspid valve during 2002-2008. Of those 416,277 patients, 19,730 had the surgery for infective endocarditis.
The investigators randomly selected 70% of those cases to develop the scoring system, reserving the remaining 30% to test it. They then developed two separate scoring systems, one intended to predict mortality alone and the other intended to predict a composite of deep external wound infection, mortality and serious morbidity, prolonged ventilation, postoperative stroke, pneumonia, renal failure, dialysis, multisystem organ failure, and readmission within 30 days of surgery.
The patients' average age was 55 years, and 67% were male. Preoperatively, 52% had active endocarditis, 23% were in renal failure, 21% had prior valve surgery, 19% had arrhythmia, and 19% had prior cerebrovascular disease.
The surgery was urgent for 50% of the patients, elective for 43%, and emergent for just under 7%. Fewer than 1% of the patients had salvage surgery.
Overall, 8.2% of the patients died, a lower percentage than that seen in other studies. But 53% had postoperative complication such as prolonged ventilation (28%), strokes within 72 hours (3%), and transient neurological deficits (1%).
After conducting a multiple logistic regression controlling for relevant demographic and clinical characteristics, the investigators used the top 13 significant risk factors to develop their scoring systems (see box). The top two risk factors were the same in the mortality and mortality/morbidity scoring systems: Patients who were emergent, salvage, or in cardiogenic shock were three times more likely to die or to have major morbidity than was the average patient. And patients in renal failure were twice as likely to experience adverse outcomes.
Dr. Gaca explained using the example of a 65-year-old man with active mitral valve endocarditis, NYHA class IV heart failure, type 1 diabetes, serum creatinine of 2.2 mg/dL, and chronic obstructive pulmonary disease. His risk score for major morbidity and mortality would be 36, and his score for mortality alone would be 35.
When those values are plotted on risk graphs, a score of 35 translates to an 11% chance of mortality, and a score of 36 translates to about a 65% chance of mortality or major morbidity.
In commenting on the study, Dr. James M. Douglas of St. Joseph's Hospital in Bellingham, Wash., noted that most of the identified risk factors were functional or physiologic, whereas in his experience anatomical peculiarities such as aneurysms and fistulae present the most vexing challenges in these patients.
Dr. Gaca acknowledged that anatomic issues do affect a patient's risk, and those factors are not included in the STS database. Another limitation of that database is the lack of microbiological data.
Disclosures: The investigators had no relevant disclosures.
Source Elsevier Global Medical News