The lack of diagnostic value of the indium scan in acute bacterial endocarditis

Abstract

Diagnosis of acute bacterial endocarditis should be made in a prompt and effective manner to institute appropriate therapy and minimize morbidity and mortality. The diagnostic criteria consist of positive blood cultures with organisms known to cause infection, echocardiographic evidence of valvular/endocardial involvement (i.e., vegetations), and a new or changing murmur (1Cunha B.A. Gill M.V. Lazar J. Acute infective endocarditis.Infect Dis Clin North Am. 1996; 10: 811-834Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar).Presumptive diagnosis is usually made by a high-grade bacteremia (i.e., 4/4 or 3/4 sets of positive blood cultures), an elevated white blood cell count and erythrocyte sedimentation rate, fever of 102°F or higher, and vegetations on echocardiography. The blood cultures are the most important laboratory test and are positive in approximately 95% of cases (2Saccente M. Cobbs C.G. Clinical approach to infective endocarditis.CardiolClin. 1996; 14: 351-362Scopus (17) Google Scholar).Echocardiography has been the most widely used imaging modality in the diagnosis of infective endocarditis. It is used to determine the presence/size of the vegetation, specific location of the vegetation, degree of valve incompetency, and complicating factors such paravalvular abscesses.Studies have shown that for the identification of vegetations in infective endocarditis, the sensitivities are 50% for transthoracic echocardiography and 90% for transesophagealechocardiography, making transesophageal echocardiography the preferred modality when clinically feasible (3Reynolds H.R. Jagen M.A. Tunick P.A. Kronzon I. Sensitivityof transthoracic versus transesophageal echocardiography for the detection of native valve vegetations in the modern era.J Am Soc Echocardiogr. 2003; 16: 67-70Abstract Full Text Full Text PDF PubMed Scopus (131) Google Scholar, 4Erbel R. Rohmann S. Drexler M. et al.Improved diagnostic value of echocardiography in patients with infective endocarditis by transesophageal approach: a prospective study.EurHeartJ. 1988; 9: 43-53Google Scholar).Indium-labeled white blood cell body scans have also been used as part of the diagnostic workup of infective endocarditis. Studies have suggested that the indium scan is most useful in detecting valvular vegetations in cases where the vegetation size is greater than 5 mm, but insensitive for vegetations less than 5 mm (5McAfee J.G. Samin A. In-111-labeled leukocytes: a review of problems in image interpretation.Radiology. 1985; 155: 221-229Crossref PubMed Scopus (81) Google Scholar, 6Borst U. Becker W. Maisch B. et al.Clinical and prognostic effect of a positive granulocyte scan in infective endocarditis.Clin Nucl Med. 1993; 18: 35-39Crossref PubMed Scopus (14) Google Scholar, 7Riba A.L. Thakur M.L. Gottschalk A. et al.Imaging experimental infective endocarditis with Indium-111-labeled blood cellular elements.Circulation. 1979; 59: 336-343Crossref PubMed Scopus (34) Google Scholar, 8Melvin E.T. Berger M. Lutzker L.G. et al.Noninvasive methods for detection of valve vegetations in infective endocarditis.Am J Cardiol. 1981; 47: 271-277Abstract Full Text PDF PubMed Scopus (56) Google Scholar, 9McDougall I.R. Baumer J.E. Lantieri R.L. Evaluation of 111In leukocyte whole body scanning.AJR Am J Roentgenol. 1979; 133: 849-854Crossref PubMed Scopus (70) Google Scholar).We reviewed our experience with indium scans as part of the diagnostic workup of non–intravenous drug abuse patients with acute bacterial endocarditis. During a 2-year period, there were 39 cases of acute bacterial endocarditis in our 600-bed university-affiliated community teaching hospital.We reviewed the records for non–intravenous drug abuse patientsdiagnosed with acute bacterial endocarditis for the past 2 years. Someof the inclusion criteria included acute febrile illness with feversgreater than 102°F and without other obvious sources; high-gradebacteremia with pathogens known to cause acute bacterial endocarditis;vegetation present on transesophageal echocardiography. Of 39 patientswith a presumptive diagnosis, 11 had vegetations on transesophagealechocardiography. Seven of these patients (mean age, 66 years; range,46 to 82 years; 4 men and 3 women) had an indium scan as part of theirworkup, satisfying all of the inclusion categories. The bacterialisolates included four staphylococcal species, two streptococci, andone Serratia. Vegetation size varied from lessthan 5 mm (2 patients) to 15 mm (2 patients). Transesophagealechocardiography revealed that 5 patients had vegetations greater than5 mm (71%). One patient had a prosthetic valve acuteendocarditis. All indium scans were negative independent of vegetationsize(Table).TableCharacteristics of the Cases of Infective EndocarditisPatientAge (years)SexIsolateValveVegetation Size (mm)146MMethicillin-sensitiveAortic15Staphylococcus aureus261FStreptococcus viridansAortic15353MMethicillin-sensitiveMitral14Staphylococcus aureus482FCoagulase-negativeAortic (prosthetic)<5Staphylococcus575MCoagulase-negativeMitral7.4Staphylococcus667FSerratiaAortic4767MStreptococcus pneumoniaeAortic6F = female; M = male. Open table in a new tab Indium scans involve thelabeling of isolated polymorphonuclear cells with 111In to observeuptake in areas of infection or inflammation(5McAfee J.G. Samin A. In-111-labeled leukocytes: a review of problems in image interpretation.Radiology. 1985; 155: 221-229Crossref PubMed Scopus (81) Google Scholar, 6Borst U. Becker W. Maisch B. et al.Clinical and prognostic effect of a positive granulocyte scan in infective endocarditis.Clin Nucl Med. 1993; 18: 35-39Crossref PubMed Scopus (14) Google Scholar, 7Riba A.L. Thakur M.L. Gottschalk A. et al.Imaging experimental infective endocarditis with Indium-111-labeled blood cellular elements.Circulation. 1979; 59: 336-343Crossref PubMed Scopus (34) Google Scholar, 8Melvin E.T. Berger M. Lutzker L.G. et al.Noninvasive methods for detection of valve vegetations in infective endocarditis.Am J Cardiol. 1981; 47: 271-277Abstract Full Text PDF PubMed Scopus (56) Google Scholar, 9McDougall I.R. Baumer J.E. Lantieri R.L. Evaluation of 111In leukocyte whole body scanning.AJR Am J Roentgenol. 1979; 133: 849-854Crossref PubMed Scopus (70) Google Scholar).Valvular vegetations greater than 5 mm are detected on indium scan(3Reynolds H.R. Jagen M.A. Tunick P.A. Kronzon I. Sensitivityof transthoracic versus transesophageal echocardiography for the detection of native valve vegetations in the modern era.J Am Soc Echocardiogr. 2003; 16: 67-70Abstract Full Text Full Text PDF PubMed Scopus (131) Google Scholar, 4Erbel R. Rohmann S. Drexler M. et al.Improved diagnostic value of echocardiography in patients with infective endocarditis by transesophageal approach: a prospective study.EurHeartJ. 1988; 9: 43-53Google Scholar).Our experience shows that even when the vegetation size was greaterthan 5 mm, the indium scan wasnegative.The diagnosis of acutebacterial endocarditis is based on history, physical examination,high-grade bacteremia with known pathogens causing acute bacterialendocarditis, acute febrile illness with fever greater than 102°F,and vegetations on echocardiography. In the era of managed health care,the average medical cost for a hospitalized case of endocarditis isapproximately $47,200(10Rubin R.J. Harrington C.A. Poon A. et al.The economic impact of Staphylococcus aureus infection in New York City hospitals.Emerg Infect Dis. 1999; 5: 9-17Crossref PubMed Scopus (379) Google Scholar).The average cost of an indium scan at our institution is$2300.We conclude that the indium scan adds unnecessary cost and does not contribute to thediagnostic workup of acute bacterial endocarditis. Further, indiumscans, in our experience, were falsely negative even with vegetationslarger than 5mm. Diagnosis of acute bacterial endocarditis should be made in a prompt and effective manner to institute appropriate therapy and minimize morbidity and mortality. The diagnostic criteria consist of positive blood cultures with organisms known to cause infection, echocardiographic evidence of valvular/endocardial involvement (i.e., vegetations), and a new or changing murmur (1Cunha B.A. Gill M.V. Lazar J. Acute infective endocarditis.Infect Dis Clin North Am. 1996; 10: 811-834Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar). Presumptive diagnosis is usually made by a high-grade bacteremia (i.e., 4/4 or 3/4 sets of positive blood cultures), an elevated white blood cell count and erythrocyte sedimentation rate, fever of 102°F or higher, and vegetations on echocardiography. The blood cultures are the most important laboratory test and are positive in approximately 95% of cases (2Saccente M. Cobbs C.G. Clinical approach to infective endocarditis.CardiolClin. 1996; 14: 351-362Scopus (17) Google Scholar). Echocardiography has been the most widely used imaging modality in the diagnosis of infective endocarditis. It is used to determine the presence/size of the vegetation, specific location of the vegetation, degree of valve incompetency, and complicating factors such paravalvular abscesses.Studies have shown that for the identification of vegetations in infective endocarditis, the sensitivities are 50% for transthoracic echocardiography and 90% for transesophagealechocardiography, making transesophageal echocardiography the preferred modality when clinically feasible (3Reynolds H.R. Jagen M.A. Tunick P.A. Kronzon I. Sensitivityof transthoracic versus transesophageal echocardiography for the detection of native valve vegetations in the modern era.J Am Soc Echocardiogr. 2003; 16: 67-70Abstract Full Text Full Text PDF PubMed Scopus (131) Google Scholar, 4Erbel R. Rohmann S. Drexler M. et al.Improved diagnostic value of echocardiography in patients with infective endocarditis by transesophageal approach: a prospective study.EurHeartJ. 1988; 9: 43-53Google Scholar). Indium-labeled white blood cell body scans have also been used as part of the diagnostic workup of infective endocarditis. Studies have suggested that the indium scan is most useful in detecting valvular vegetations in cases where the vegetation size is greater than 5 mm, but insensitive for vegetations less than 5 mm (5McAfee J.G. Samin A. In-111-labeled leukocytes: a review of problems in image interpretation.Radiology. 1985; 155: 221-229Crossref PubMed Scopus (81) Google Scholar, 6Borst U. Becker W. Maisch B. et al.Clinical and prognostic effect of a positive granulocyte scan in infective endocarditis.Clin Nucl Med. 1993; 18: 35-39Crossref PubMed Scopus (14) Google Scholar, 7Riba A.L. Thakur M.L. Gottschalk A. et al.Imaging experimental infective endocarditis with Indium-111-labeled blood cellular elements.Circulation. 1979; 59: 336-343Crossref PubMed Scopus (34) Google Scholar, 8Melvin E.T. Berger M. Lutzker L.G. et al.Noninvasive methods for detection of valve vegetations in infective endocarditis.Am J Cardiol. 1981; 47: 271-277Abstract Full Text PDF PubMed Scopus (56) Google Scholar, 9McDougall I.R. Baumer J.E. Lantieri R.L. Evaluation of 111In leukocyte whole body scanning.AJR Am J Roentgenol. 1979; 133: 849-854Crossref PubMed Scopus (70) Google Scholar). We reviewed our experience with indium scans as part of the diagnostic workup of non–intravenous drug abuse patients with acute bacterial endocarditis. During a 2-year period, there were 39 cases of acute bacterial endocarditis in our 600-bed university-affiliated community teaching hospital. We reviewed the records for non–intravenous drug abuse patientsdiagnosed with acute bacterial endocarditis for the past 2 years. Someof the inclusion criteria included acute febrile illness with feversgreater than 102°F and without other obvious sources; high-gradebacteremia with pathogens known to cause acute bacterial endocarditis;vegetation present on transesophageal echocardiography. Of 39 patientswith a presumptive diagnosis, 11 had vegetations on transesophagealechocardiography. Seven of these patients (mean age, 66 years; range,46 to 82 years; 4 men and 3 women) had an indium scan as part of theirworkup, satisfying all of the inclusion categories. The bacterialisolates included four staphylococcal species, two streptococci, andone Serratia. Vegetation size varied from lessthan 5 mm (2 patients) to 15 mm (2 patients). Transesophagealechocardiography revealed that 5 patients had vegetations greater than5 mm (71%). One patient had a prosthetic valve acuteendocarditis. All indium scans were negative independent of vegetationsize(Table). F = female; M = male. Indium scans involve thelabeling of isolated polymorphonuclear cells with 111In to observeuptake in areas of infection or inflammation(5McAfee J.G. Samin A. In-111-labeled leukocytes: a review of problems in image interpretation.Radiology. 1985; 155: 221-229Crossref PubMed Scopus (81) Google Scholar, 6Borst U. Becker W. Maisch B. et al.Clinical and prognostic effect of a positive granulocyte scan in infective endocarditis.Clin Nucl Med. 1993; 18: 35-39Crossref PubMed Scopus (14) Google Scholar, 7Riba A.L. Thakur M.L. Gottschalk A. et al.Imaging experimental infective endocarditis with Indium-111-labeled blood cellular elements.Circulation. 1979; 59: 336-343Crossref PubMed Scopus (34) Google Scholar, 8Melvin E.T. Berger M. Lutzker L.G. et al.Noninvasive methods for detection of valve vegetations in infective endocarditis.Am J Cardiol. 1981; 47: 271-277Abstract Full Text PDF PubMed Scopus (56) Google Scholar, 9McDougall I.R. Baumer J.E. Lantieri R.L. Evaluation of 111In leukocyte whole body scanning.AJR Am J Roentgenol. 1979; 133: 849-854Crossref PubMed Scopus (70) Google Scholar).Valvular vegetations greater than 5 mm are detected on indium scan(3Reynolds H.R. Jagen M.A. Tunick P.A. Kronzon I. Sensitivityof transthoracic versus transesophageal echocardiography for the detection of native valve vegetations in the modern era.J Am Soc Echocardiogr. 2003; 16: 67-70Abstract Full Text Full Text PDF PubMed Scopus (131) Google Scholar, 4Erbel R. Rohmann S. Drexler M. et al.Improved diagnostic value of echocardiography in patients with infective endocarditis by transesophageal approach: a prospective study.EurHeartJ. 1988; 9: 43-53Google Scholar).Our experience shows that even when the vegetation size was greaterthan 5 mm, the indium scan wasnegative. The diagnosis of acutebacterial endocarditis is based on history, physical examination,high-grade bacteremia with known pathogens causing acute bacterialendocarditis, acute febrile illness with fever greater than 102°F,and vegetations on echocardiography. In the era of managed health care,the average medical cost for a hospitalized case of endocarditis isapproximately $47,200(10Rubin R.J. Harrington C.A. Poon A. et al.The economic impact of Staphylococcus aureus infection in New York City hospitals.Emerg Infect Dis. 1999; 5: 9-17Crossref PubMed Scopus (379) Google Scholar).The average cost of an indium scan at our institution is$2300. We conclude that the indium scan adds unnecessary cost and does not contribute to thediagnostic workup of acute bacterial endocarditis. Further, indiumscans, in our experience, were falsely negative even with vegetationslarger than 5mm.