False Localization of Site of Endocarditis by Cardiac Catheterization with Quantitative Cultures MICHAEL BENNISH, M.D., ROBERT A. WEINSTEIN, M.D., The authors present a patient with relapsing Pseudomonas aeruginosa endocarditis in whom cardiac catheterization with quantitative cultures falsely localized the infection to the tri- cuspid valve, probably because the patient was having inter- mittent rather than continuous bacteremia. After catheteriza- tion the patient developed mitral insufficiency and congestive heart failure. This experience suggests that quantitative cul- tures during cardiac catheterization may give misleading re- sults and that the procedure may have significant complica- tions. (Key words: Cardiac catheterization; Infective endocarditis; Blood cultures) Am J Clin Pathol 1985; 83: 130-131 CARDIAC CATHETERIZATION with quantitative blood cultures from various intravascular sites is dis- cussed in standard references as a means of identifying the site of valvular involvement in selected cases of in- fective endocarditis. 56 Published experience with this procedure is limited to two patients. 3 In each of those cases, there was an increase in bacterial colony counts when cultures were obtained just distal to the infected valve. We report a patient with relapsing Pseudomonas aer- uginosa endocarditis in whom cardiac catheterization with quantitative cultures, done when noninvasive means had failed to locate the site of infection, gave misleading data and was followed by the abrupt develop- ment of mitral insufficiency. Report of a Case A 31-year-old man was admitted with a six-day history of fever and myalgia. He denied parenteral drug use. His temperature was 39.4 °C. Osier nodes and Janeway lesions were seen on his palms and soles. A grade II/VI short systolic murmur was heard at the lower left sternal border. The remainder of the examination was normal. A chest x-ray was normal. M-mode and two dimensional echocar- diograms showed no evidence of vegetations. Each of five blood cul- tures grew P. aeruginosa. We treated the patient with intravenous tobramycin and ticarcillin for six weeks. He became afebrile within 48 Received December 21, 1983; received revised manuscript and ac- cepted for publication February 1, 1984. Supported in part by an Institutional National Research Service Award from the NIAID (5T32AI07099-04) and the Home for Desti- tute Crippled Children Health Care and Medical Research Program. Address reprint requests to Dr. Weinstein: Department of Medicine, 110 Baumgarten, Michael Reese Hospital and Medical Center, Lake Shore Drive at 31st Street, Chicago, Illinois 60616. >., SHERWIN A. KABINS, M.D., AND MUKESH C. JAIN, M.D. Divisions of Infectious Diseases and Cardiology, Department of Medicine, Michael Reese Hospital and Medical Center; Section of Infectious Diseases, Department of Pediatrics, Wyler Children's Hospital; and the Pritzker School of Medicine, University of Chicago, Chicago, Illinois hours. Numerous blood cultures during therapy were sterile. Peak serum bactericidal levels were at least 1:8. At the end of therapy repeat M-mode and two-dimensional echocardiograms were normal. Twenty-four hours after we discontinued antibiotics the patient's fever recurred, and one of two blood cultures grew P. aeruginosa with the same antibiogram and serotype as his previous isolates. He was restarted on tobramycin and ticarcillin and was afebrile within 24 hours. The inability to localize the site of endocarditis by noninvasive means led us to perform cardiac catheterization with quantitative cul- tures. Methods Antibiotics were withheld for 48 hours. At catheteri- zation, under direct image intensification, samples for cultures were obtained sequentially, first from the venous and pulmonary circulation, followed by the sys- temic circulation (with a 7F pigtail catheter). Sampling of each side was then repeated. Twenty samples each containing 4-5 mL of blood in sterile heparinized sy- ringes were obtained. The specimens immediately were placed on ice and cultured within 30 minutes. Each specimen was divided intofivealiquots. Two 1.0-mL and two 0.5-mL aliquots each were mixed with 6 mL of molten nutrient agar (Difco, Detroit, MI) at 48 °C and poured into a petri dish to make a thin suspension. Another 1 mL was added to 30 mL of Columbia broth (BBL, Cockeysville, MD). Bacterial colonies were counted at 24 and 48 hours, and broths routinely were subcultured after 24 and 48 hours and seven days of incubation at 36 °C. Results Only 2 of the first 14 specimens had growth (Table 1). Starting with the 15th specimen, drawn from the right ventricle, all samples grew Pseudomonas. That speci- men had the highest colony counts; the subsequent spec- imens showed a steady decrement. There was no visual or hemodynamic evidence of valvular abnormalities or insufficiency on aortic root or ventricular angiograms. 130 Downloaded from https://academic.oup.com/ajcp/article-abstract/83/1/130/1792434 by guest on 28 July 2018