- Email: [email protected]
Are high-velocity tricuspid and pulmonary regurgitation endocarditis risk substrates?
Are high-velocity tricuspid and pulmonary regurgitation endocarditis risk substrates? Hidemi Dodo, MD, Joseph K. Perloff, MD, John S. Child, MD, Pamela D. Miner, RN, MN, and David A. Pegues, MD Los Angeles, Calif.
Background A major predisposing cause of infective endocarditis is a susceptible cardiac substrate characterized by high-velocity turbulent flow. However, the risk incurred by high-pressure, high-velocity regurgitation across inherently normal pulmonary and tricuspid valves has not hitherto been examined.
Methods and Results This study focused on 186 adult patients with congenital heart disease who had pulmonary vascular disease and inherently normal right-sided pulmonary and tricuspid valves. The observation period was approximately 1646 patient-years. Exclusion criteria were coexisting lesions that might have served as independent risk substrates for infective endocarditis. High-velocity turbulent pulmonary and tricuspid regurgitation were identified and quantified by color flow imaging and continuous wave Doppler echocardiography. Diagnoses of infective endocarditis were based on established clinical and laboratory criteria. Tricuspid regurgitation was moderate to severe in 80 patients and mild or absent in 106 patients. Pulmonary regurgitation was moderate to severe in 84 patients and mild or absent in 102 patients. With the exception of a single habitual intravenous drug abuser, no patient, irrespective of the degree of high-velocity turbulent pulmonary or tricuspid regurgitation, had infective endocarditis.
Conclusions High-velocity turbulent flow across inherently normal pulmonary and tricuspid valves rendered incompetent by pulmonary hypertension may represent a relatively low-risk or no-risk substrate for infective endocarditis. (Am Heart J 1998;136:109-14.)
Two major risk factors are believed to predispose to infective endocarditis: a susceptible cardiac or vascular substrate and bacteremia.1-4 Susceptible substrates are characterized by high-velocity turbulent flow that causes a focal increase in shear stress and a jet impact that causes endothelial damage.1-4 Susceptibility has been classified into three categories: (1) high risk, (2) moderate risk, and (3) low or no risk (negligible, that is, risk no greater than the general population).1,3,4 We sought to categorize the risk of infective endocarditis on inherently normal right-sided tricuspid and pulmonary valves rendered incompetent by pulmonary hypertension. This clinically relevant issue has hitherto not been addressed.
From the Divisions of Cardiology and Infectious Disease, Departments of Medicine and Pediatrics, and the UCLA Adult Congenital Heart Disease Center, University of California, Los Angeles. Submitted Aug. 20, 1997; accepted Feb. 6, 1998. Reprint requests: Joseph K. Perloff, MD, Division of Cardiology, Room 47-123 CHS, UCLA Center for the Health Sciences, Los Angeles, CA 90095-1679. Copyright © 1998 by Mosby, Inc. 0002-8703/98/$5.00 + 0 4/1/89397
Methods Of 1360 patients entered into the UCLA Adult Congenital Heart Disease Center Registry from 1980 through 1997, 222 had pulmonary vascular disease. Of those 222 patients, this study comprised 186 with tricuspid and pulmonary valves that were inherently normal apart from the regurgitation incurred by high pulmonary vascular resistance. Because potential susceptibility of incompetent tricuspid and pulmonary valves to infective endocarditis may differ, these two valves were considered separately (Tables I, II, and III). Patients were aged 22 years to 68 years (mean ± SD, 39 ± 9 years). Seventy-eight were men and 108 were women. None had coexisting malformations that might have served as independent substrates for infective endocarditis. It should be emphasized that reversed (right-to-left) shunts at ventricular or great arterial level are characterized by low-velocity nonturbulent flow, no jet impact site, and no risk of infective endocarditis.4-6 Accordingly, high-velocity pulmonary and tricuspid regurgitation were the only potential risk substrates. Neither antibiotic nor nonantibiotic prophylaxis against infective endocarditis had been recommended,4 and no patient had a history of a protracted febrile illness of unknown cause. Six other patients had received intermittent antibiotic prophylaxis and were therefore excluded. All patients had noninverted ventricles and nontransposed great arteries, so
American Heart Journal July 1998
110 Dodo et al.
Table I. High-velocity regurgitation: Absent or mild Diagnosis VSD ASD PDA PPH APW TAPVC Total
No. Pulmonary 63 22 3 14 0 0 102
No. Tricuspid 65 22 3 14 1 1 106
VSD, Ventricular septal defect; ASD, atrial septal defect; PDA, patent ductus arteriosus; PPH, primary pulmonary hypertension; APW, aortopulmonary window; TAPVC, total anomalous pulmonary venous connection, supradiaphragmatic.
Table II. High-velocity pulmonary regurgitation: Moderate to severe Diagnosis
No.
VSD ASD PDA VSD/ASD VSD/PDA PPH APW TAPVC Total
35 27 3 4 1 12 1 1 84
Abbreviations as in Table I.
Table III. High-velocity tricuspid regurgitation: Moderate to severe Diagnosis
No.
VSD ASD PDA VSD/ASD VSD/PDA PPH APW TAPVC Total
32 24 4 5 3 12 0 0 80
Abbreviations as in Table I.
the incompetent pulmonary and tricuspid valves resided in the right side of the heart. The patients were reevaluated in our clinic by the coauthors at intervals of 6 months or less. Follow-up ranged from 2 years to 17 years (average 8.8 years). The observation period comprised approximately 1640 patient-years. Patients
who were not reevaluated by us within 1 year (generally those residing outside our area) were contacted, as were their physicians. The patients and physicians responded to a questionnaire designed to disclose the lowest probability of infective endocarditis. The questionnaire was modeled after the questionnaire used in the second Natural History Study of Congenital Heart Defects.7 Fourteen patients were lost to follow-up after observation periods of 5 to 11 years. Forty-one patients (24%) died during the course of observation. Necropsies were performed in 17 patients by the same cardiac pathologist. Two-dimensional transthoracic echocardiography with color flow imaging and spectral Doppler interrogation were performed in all patients, usually on several occasions. Twelve patients had transesophageal echocardiograms for reasons other than suspected infective endocarditis. Echocardiography established or confirmed the diagnosis of the basic congenital cardiac malformation and excluded high-velocity turbulent flow at sites other than the pulmonary or tricuspid valve. The pulmonary and tricuspid valves were examined for morphologic abnormalities by echocardiography in all cases and on gross examination in the 17 necropsy cases. The presence and degree of pulmonary and tricuspid regurgitation were determined by color flow imaging by the use of established criteria.8-10 For the purpose of this study, the degree of pulmonary and tricuspid regurgitant flow was characterized as absent or mild and moderate to severe. Peak velocity of regurgitant flow was based on color flow–guided, continuous wave Doppler echocardiography.8-11 Pulmonary arterial diastolic pressure and right ventricular systolic pressure were elevated by definition because of pulmonary vascular disease. The minimum pulmonary regurgitant end-diastolic velocity of 3.2 m/sec and the minimum tricuspid regurgitant peak systolic velocity of 4.7 m/sec established that regurgitant flow velocity was well above normal.8-11 In addition to the echocardiographic criteria, Graham Steele murmurs of pulmonary hypertensive pulmonary regurgitation and high-frequency holosystolic murmurs of high-pressure tricuspid regurgitation were routinely sought. Coexisting aortic regurgitation was excluded by Doppler echocardiography with color flow imaging. Criteria for the clinical diagnosis of infective endocarditis included an appropriate febrile illness, appropriate physical signs, positive blood cultures,1,3,4,12,13 and convincing evidence of a vegetation on transthoracic or transesophageal echocardiography.14,15 Analogous data were assembled on the 1174 UCLA Adult Congenital Heart Disease Registry patients not included in the above formal study. Forty-five percent were women and 55% were men. Ages ranged from 18 years to 95 years (mean ± SD, 42 ± 11 years). Follow-up ranged from 1 year to 20 years (mean ± SD, 11.1 ± 3.9 years). The observation period comprised approximately 11,860 patient-years. Seventy-two patients were lost to follow-up or died. The infective endo-
American Heart Journal Volume 136, Number 1
Figure 1
Necropsy specimen from 38-year-old woman with nonrestrictive ventricular septal defect and Eisenmenger reaction. Arrows point to nodular thickening along high-pressure closure line of inherently normal tricuspid valve (TV).
carditis questionnaire used for the pulmonary vascular disease study group (see earlier) was not used. The relative proportion of patients developing infective endocarditis and the number of episodes per 1000 patientyears were determined for the two patient groups, that is, the pulmonary vascular disease study group and the remaining patients in the registry.16-19 Exact 95% confidence intervals (95% CI) were calculated, and the significance of rate ratios was determined by using EpiInfo computer software (version 6.02, Centers for Disease Control, Atlanta, Ga.). Probability values are two tailed.
Results The tricuspid valves had inherently normal structure on the basis of echocardiography in all of the 186 study patients and on the basis of necropsy examination in 17 patients. Fifteen of the 17 necropsy specimens had nodular changes along the tricuspid leaflet closure lines that were considered secondary responses to chronic high-pressure tricuspid valve closure (Fig. 1). Despite high-velocity turbulent regurgitant flow, a jet impact site could not be identified in the 17 necropsy specimens. Tricuspid regurgitation was mild or absent in 106 patients (Table I) and moderate to severe in 80 (Table III). Tricuspid regurgitant peak systolic velocity ranged from 4.7 to 5.8 m/sec. No patient had documented infective endocarditis irre-
Dodo et al. 111
Figure 2
Necropsy specimen from 54-year-old woman with nonrestrictive ventricular septal defect (VSD) and Eisenmenger reaction. Paired bold arrows point to nodular thickening along high-pressure closure line of inherently normal pulmonary valve (PV).
spective of the presence or degree of high-velocity turbulent regurgitant flow. The pulmonary valves had inherently normal structure on the basis of echocardiography in all 186 patients and on the basis of necropsy examination in 17 patients. Despite high-pressure closure, only 1 of the 17 pulmonary valves examined at necropsy had secondary changes (Fig. 2), and despite high-velocity turbulent regurgitant flow, a jet impact site could not be identified in the 17 necropsy specimens. Pulmonary regurgitation was mild to absent in 102 patients (Table I) and moderate to severe in 84 (Table II). Pulmonary regurgitant end-diastolic velocities ranged from 3.2 to 3.8 m/sec. Infective endocarditis was identified in 1 patient, a 33-year-old cyanotic woman with pulmonary vascular disease and a nonrestrictive ostium secundum atrial septal defect, who developed Staphylococcus aureus endocarditis on the pulmonary valve while actively engaged in intravenous drug abuse. Transthoracic echocardiography identified a mobile vegetation attached to a moderately incompetent pulmonary valve. In a previous echocardiogram, the pulmonary valve was considered structurally normal and mildly incompetent.
112 Dodo et al.
Assuming that pulmonary valve infective endocarditis in the one patient was due to risk incurred by pulmonary vascular disease rather than to risk incurred by intravenous drug abuse on an inherently normal pulmonary valve, the rate of infective endocarditis in the primary study group was 0.54% (1 of 186) of patients or 0.61 episodes (1 of 640) per 1000 patient-years (95% CI 0.0 to 3.39). In the 1174 patients with congenital heart disease who were not included in the study group, 76 patients experienced 85 episodes of infective endocarditis. One patient experienced five episodes, and five patients experienced two episodes each. Thus the rate of infective endocarditis was 6.47% (76 of 1174) of patients or 7.17 episodes (85 of 11,860) per 1000 patient-years. On the basis of these estimates, the cumulative rate of infective endocarditis among the 1174 patients was almost 12 times greater than the rate among the study patients with pulmonary vascular disease (rate ratio 11.75; 95% CI 1.64 to 84.35; p = 0.002). Among the 76 patients who experienced infective endocarditis, 35 were unoperated (46%) and 41 were postoperative (54%). In the 35 unoperated patients, the diagnostic categories that served as the substrates for infective endocarditis included in relative order of frequency: tetralogy of Fallot, 24% (8% with pulmonary stenosis, 16% with pulmonary atresia); doubleoutlet right ventricle with pulmonary stenosis, 12%; restrictive ventricular septal defect, 12%; tricuspid atresia with normally related great arteries and subpulmonary stenosis, 10%; bicuspid aortic valve, 10%; congenitally corrected transposition of the great arteries, 8% (5% with pulmonary stenosis, 3% with pulmonary atresia); univentricular heart (left ventricular morphology) with pulmonary stenosis, 5%; and all other categories single examples, 21%. Of the 41 postoperative patients who had infective endocarditis, 22 had undergone palliative procedures (systemic to pulmonary arterial shunts, especially Blalock-Taussig). Nineteen of the 41 postoperative patients had reparative surgery for complex cyanotic congenital heart disease leaving behind residua and sequelae, including conduits and prosthetic valves.
Discussion High-velocity turbulent flow is regarded as a major predisposing cause of infective endocarditis either at the site generating the high-velocity turbulence (increased rate of shear) or at a jet impact site.1-4 Susceptibility at either site is believed to depend on endothelial damage, with risk determined by the
American Heart Journal July 1998
degree of damage to endothelial surfaces and by the pathogenicity (virulence) of the microorganism.3,4 The degree of endothelial damage assumes more importance when the microorganism is of low virulence and the bacteremia is transient and assumes less importance when the microorganism is of high virulence and the bacteremia is sustained or recurrent.3 Aortic regurgitation and mitral regurgitation are considered risk substrates for infective endocarditis because of high-velocity turbulent regurgitant flow.1,3,4 High-pressure pulmonary regurgitation (high-velocity turbulent diastolic flow) or high-pressure tricuspid regurgitation (high-velocity turbulent systolic flow) might be regarded as risk substrates analogous to aortic and mitral regurgitation. Inherently normal pulmonary valves are often rendered incompetent by pulmonary hypertension, but inherently normal trileaflet aortic valves typically remain competent even in the presence of substantial systemic hypertension. Such aortic valves have not been considered susceptible to infective endocarditis. However, in tetralogy of Fallot, especially with pulmonary atresia, initially competent trileaflet aortic valves housed in an enlarged aortic root accrue aortic regurgitation decade by decade of adult life.20 Among the 76 nonstudy patients who experienced infective endocarditis, 24% had tetralogy of Fallot, 8% with pulmonary stenosis and 6% with pulmonary atresia. The incidence of aortic regurgitation was highest in the latter category. Interestingly, inherently normal mitral valves are frequently rendered incompetent because left ventricular dilatation (heart failure or dilated cardiomyopathy, for example) results in lateral migration of the papillary muscles and because depressed systolic function prevents adequate reduction in mitral annular circumference.21 Infective endocarditis on these incompetent mitral valves is not known to incur insofar as we could determine, although susceptibility has not been systematically examined. Apart from intravenous drug abuse associated with sustained, recurrent bacteremia with high-virulence S. aureus, infective endocarditis rarely involves cardiac valves that are normal in structure and function.3,12,22 There is a unique propensity for intravenous drug abuse infective endocarditis to involve normal tricuspid and pulmonary valves, although normal left-sided valves are by no means spared.22,23 A pivotal question in our study is whether the single patient who incurred S. aureus infective endocarditis during active intravenous drug abuse was an example of infection
American Heart Journal Volume 136, Number 1
on a structurally and functionally normal pulmonary valve or whether pulmonary hypertension contributed to or caused the infection by lowering the threshold to sustained bacteremia with a high-virulence organism. Importantly, a first episode of infectious endocarditis has a far grater propensity to involve the tricuspid valve (44%) than the pulmonary valve (3%).22 If it is assumed that the single example of infective endocarditis in our study group was related to chronic high-pressure closure of a pulmonary valve that was judged by echocardiography to have been structurally normal and mildly incompetent before infection, then the cumulative rate of infective endocarditis was 0.061% or 0.61 per 1000 patient-years (95% CI 0.0 to 3.39). Conversely, if it is assumed that this single case was an example of right-sided infective endocarditis unrelated to pulmonary hypertension, then the rate of infection was 0 per 1640 patient-years in our study population (95% CI 0.0 to 2.25). Prolonged high-pressure closure associated with pulmonary hypertension can induce morphologic changes on tricuspid or pulmonary valves of inherently normal structure. In our necropsy patients, minor thickening of the tricuspid valve closure line was commonly so related (Fig. 1), but analogous changes were found on only one pulmonary valve (Fig. 2) despite the probability of inherent pulmonary cuspal inequality. Assuming the presence of these structural changes, infective endocarditis did not occur in any of the 186 patients apart from the single case of intravenous drug abuse irrespective of the presence or degree of high-velocity tricuspid or pulmonary regurgitation. Nor could we find in the literature an example of infective endocarditis involving tricuspid or pulmonary valves that were inherently normal but exposed to high-velocity pulmonary hypertensive regurgitant flow. On the basis of our necropsy observations, highvelocity regurgitation across tricuspid or pulmonary valves of inherently normal structure apparently causes little or no increase in shear rate at valve level and little or no jet impact on endothelial surfaces remote from these valves. Accordingly, endothelial damage—a sine qua non of susceptibility3,4—may be insignificant or absent. When considering rare events such as the incidence of infective endocarditis, risk can be estimated by comparing cases (herein the pulmonary vascular disease study patients) per 1000 patient-years with the cases of infective endocarditis per 1000 patient-years in the general population.16-19 To accumulate sufficient
Dodo et al. 113
patient-years to validate such a study would require a multicenter collaboration. However, we should not lose sight of the fact that there is a difference between years of observation derived from large numbers of patients observed for short periods of time versus smaller numbers observed for long periods of time, as in our study. Alternatively, infective endocarditis risk might be estimated by comparing our pulmonary vascular disease study patients (186 of 1360) with the additional patients (1174 of 1360) in the UCLA Adult Congenital Heart Disease Registry. Because a substantial portion of the 1174 patients had been instructed regarding prophylaxis for infective endocarditis, and because none of the pulmonary vascular disease study patients has been so advised, a comparison would provide the maximal rather than the minimal estimate of relative risk.
Limitations of the study and clinical implications regarding prophylaxis When considering rare events such the occurrence of infective endocarditis, the small size of our study is a limitation, although the long duration of follow-up is a strength. To resolve questions of relative risks that lie in the range of 0 to 1 event per 1000 patient-years would require a multicenter study involving large numbers of patients. We chose instead to report our clinical/epidemiologic observations without reaching a final conclusion regarding risk of infective endocarditis on inherently normal tricuspid and pulmonary valves rendered incompetent by pulmonary hypertension. During approximately 1640 patient-years of observation, we identified 1 case of infective endocarditis on the pulmonary valve of a habitual intravenous drug abuser. If this case is included, the cumulative rate of infective endocarditis was 0.061% or 0.61 per 1000 patient-years (95% CI 0.0 to 3.39 per 1000 patientyears). In the general adult population of the United States, the annual incidence of infective endocarditis is reportedly 0.005%, or 1 case in 20,000 persons per year.16-18 In an urban European population, the annual incidence is reportedly 0.006%, or 1.2 cases in 20,000 persons per year.19 Analysis of the 1174 additional Adult Congenital Heart Disease Registry patients who were apart from those in our study population disclosed 85 episodes of infective endocarditis among 76 patients during an observation period of approximately 11,860 patientyears. Given the rate of infective endocarditis in the 186 pulmonary vascular disease study patients versus
American Heart Journal July 1998
114 Dodo et al.
the rate in the 1174 additional registry patients, the rate was 10.5 to 12 times higher in the latter group. Our clinical and epidemiologic observations leave open the question of whether high-velocity turbulent flow across inherently normal tricuspid or pulmonary valves rendered incompetent by pulmonary hypertension incurs a risk of infective endocarditis. These substrates may be low risk or no risk.24 However, until this clinically important question is satisfactorily resolved, prophylaxis against infective endocarditis remains advisable.
References 1. Dajani AS, Taubert KA, Wilson W, Bolger AF, Bayer A, Ferrieri P, et al. Prevention of bacterial endocarditis: recommendations by the American Heart Association. Circulation 1997;96:358-66. 2. Rodbard S. Blood velocity and endocarditis. Circulation 1963;27:18-28. 3. Steckelberg JM, Wilson WR. Risk factors for infective endocarditis. Infect Dis Clin North Am 1993;7:9-19. 4. Child JS, Perloff JK, Kubak B. Infective endocarditis: risks and prophylaxis. In: Perloff JK, Child JS. Congenital heart disease in adults. 2nd ed. Philadelphia: WB Saunders Co; 1998. p. 129-43. 5. Gersony WM, Hayes CJ, Driscoll DJ, Keane JF, Kidd L, et al. Bacterial endocarditis in patients with aortic stenosis, pulmonary stenosis, or ventricular septal defect. Circulation 1993;87(suppl I):I-121-6. 6. Corone P, Levy A, Hallali P, Davido A, Wyler Y, Corone A. A propos de 54 cas d’endocardites infectieuses observès en 32 ans sur une population de 2038 cardiopathies congènitales. Arch Mal Coeur 1989;82:779-84. 7. O’Fallon WM, Crowson CS, Rings LJ, Weidman WH, Driscoll DJ, et al. Second natural history study of congenital heart defects: materials and methods. Circulation 1993;87(suppl I):I-4-15. 8. Reeder GS, Currie PJ, Hagler DJ, Tajik AJ, Seward JB. Use of Doppler techniques (continuous-wave, pulse-wave, and color flow imaging) in the noninvasive hemodynamic assessment of congenital heart disease. Mayo Clin Proc 1986;61:725-44. 9. Sahn DJ. Application of color flow mapping in pediatric cardiology. Cardiol Clin 1989;7:255-64. 10. Otto CM, Pearlman AS. Valvular regurgitation: diagnoses, quan-
11. 12.
13. 14.
15.
16.
17.
18.
19.
20. 21. 22.
23. 24.
tification and clinical approach. In: Textbook of clinical echocardiography. Philadelphia: W.B. Saunders Company; 1995. p. 243-78. Schiller NB. Pulmonary artery pressure estimation by Doppler and two-dimensional echocardiography. Cardiol Clin 1990;8:277-87. Kubak BM, Nimmagadda AP, Holt CD. Advances in medical and antibiotic management of infective endocarditis. Cardiol Clin 1996; 14:405-36. Durack DT. Prevention of infective endocarditis. N Engl J Med 1995;332:38-44. Shively BK, Gurule FT, Roldan GA, Leggett JH, Schiller NB. Diagnostic value of transesophageal compared with transthoracic echocardiography in infective endocarditis. J Am Coll Cardiol 1991;18:391-7. Sanfelippo AJ, Picard MH, Newell JB, Rosas E, Davidoff R, Thomas JD, et al. Echocardiographic assessment of patients with infectious endocarditis: prediction of risk for complications. J Am Coll Cardiol 1991;18:1191-9. MacMahon SW, Roberts JK, Kramer-Fox R, Zuker DM, Roberts RB, Deuereux RB. Mitral valve prolapse and infective endocarditis. Am Heart J 1987;113:1291-8. Kozak LJ, Moien M. Detailed diagnoses and procedures for patients discharged from short-stay hospitals: United States, 1983. Hyattsville, Md.: National Center for Health Statistics; 1985, DHHS publication No. (PHS) 85-1643. (Vital and Health Statistics). Lawrence L. Detailed diagnoses and procedures for patients discharged from short-stay hospitals: United States, 1984. Hyattsville, Md.: National Center for Health Statistics; 1985, DHHS publication No. (PHS) 86-1647. (Vital and Health Statistics). Hogevik H, Alaison L, Anderson R, Lindberg J, Alestig K. Epidemiologic aspects of infective endocarditis in an urban population: a 5year prospective study. Medicine (Baltimore) 1995;74:324-39. Marelli AJ, Perloff JK, Child JS, Laks H. Pulmonary atresia with ventricular septal defect. Circulation 1994;89:243-51. Perloff JK, Roberts WC. The mitral apparatus: functional anatomy of mitral regurgitation. Circulation 1972;46:227-39. Levine DP, Crane LR, Zervos MJ. Bacteremia in narcotic addicts at the Detroit Medical Center, II: infectious endocarditis: a prospective comparative study. Rev Infect Dis 1986;8:374-96. Dressler FA, Roberts WC. Infective endocarditis in opiate addicts: analysis of 80 cases studied at necropsy. Am J Cardiol 1989;63:1240-57. Michael PL, Acar J. Native cardiac disease predisposing to infective endocarditis. Eur Heart J 1995;16:2-6.
Background A major predisposing cause of infective endocarditis is a susceptible cardiac substrate characterized by high-velocity turbulent flow. However, the risk incurred by high-pressure, high-velocity regurgitation across inherently normal pulmonary and tricuspid valves has not hitherto been examined.
Methods and Results This study focused on 186 adult patients with congenital heart disease who had pulmonary vascular disease and inherently normal right-sided pulmonary and tricuspid valves. The observation period was approximately 1646 patient-years. Exclusion criteria were coexisting lesions that might have served as independent risk substrates for infective endocarditis. High-velocity turbulent pulmonary and tricuspid regurgitation were identified and quantified by color flow imaging and continuous wave Doppler echocardiography. Diagnoses of infective endocarditis were based on established clinical and laboratory criteria. Tricuspid regurgitation was moderate to severe in 80 patients and mild or absent in 106 patients. Pulmonary regurgitation was moderate to severe in 84 patients and mild or absent in 102 patients. With the exception of a single habitual intravenous drug abuser, no patient, irrespective of the degree of high-velocity turbulent pulmonary or tricuspid regurgitation, had infective endocarditis.
Conclusions High-velocity turbulent flow across inherently normal pulmonary and tricuspid valves rendered incompetent by pulmonary hypertension may represent a relatively low-risk or no-risk substrate for infective endocarditis. (Am Heart J 1998;136:109-14.)
Two major risk factors are believed to predispose to infective endocarditis: a susceptible cardiac or vascular substrate and bacteremia.1-4 Susceptible substrates are characterized by high-velocity turbulent flow that causes a focal increase in shear stress and a jet impact that causes endothelial damage.1-4 Susceptibility has been classified into three categories: (1) high risk, (2) moderate risk, and (3) low or no risk (negligible, that is, risk no greater than the general population).1,3,4 We sought to categorize the risk of infective endocarditis on inherently normal right-sided tricuspid and pulmonary valves rendered incompetent by pulmonary hypertension. This clinically relevant issue has hitherto not been addressed.
From the Divisions of Cardiology and Infectious Disease, Departments of Medicine and Pediatrics, and the UCLA Adult Congenital Heart Disease Center, University of California, Los Angeles. Submitted Aug. 20, 1997; accepted Feb. 6, 1998. Reprint requests: Joseph K. Perloff, MD, Division of Cardiology, Room 47-123 CHS, UCLA Center for the Health Sciences, Los Angeles, CA 90095-1679. Copyright © 1998 by Mosby, Inc. 0002-8703/98/$5.00 + 0 4/1/89397
Methods Of 1360 patients entered into the UCLA Adult Congenital Heart Disease Center Registry from 1980 through 1997, 222 had pulmonary vascular disease. Of those 222 patients, this study comprised 186 with tricuspid and pulmonary valves that were inherently normal apart from the regurgitation incurred by high pulmonary vascular resistance. Because potential susceptibility of incompetent tricuspid and pulmonary valves to infective endocarditis may differ, these two valves were considered separately (Tables I, II, and III). Patients were aged 22 years to 68 years (mean ± SD, 39 ± 9 years). Seventy-eight were men and 108 were women. None had coexisting malformations that might have served as independent substrates for infective endocarditis. It should be emphasized that reversed (right-to-left) shunts at ventricular or great arterial level are characterized by low-velocity nonturbulent flow, no jet impact site, and no risk of infective endocarditis.4-6 Accordingly, high-velocity pulmonary and tricuspid regurgitation were the only potential risk substrates. Neither antibiotic nor nonantibiotic prophylaxis against infective endocarditis had been recommended,4 and no patient had a history of a protracted febrile illness of unknown cause. Six other patients had received intermittent antibiotic prophylaxis and were therefore excluded. All patients had noninverted ventricles and nontransposed great arteries, so
American Heart Journal July 1998
110 Dodo et al.
Table I. High-velocity regurgitation: Absent or mild Diagnosis VSD ASD PDA PPH APW TAPVC Total
No. Pulmonary 63 22 3 14 0 0 102
No. Tricuspid 65 22 3 14 1 1 106
VSD, Ventricular septal defect; ASD, atrial septal defect; PDA, patent ductus arteriosus; PPH, primary pulmonary hypertension; APW, aortopulmonary window; TAPVC, total anomalous pulmonary venous connection, supradiaphragmatic.
Table II. High-velocity pulmonary regurgitation: Moderate to severe Diagnosis
No.
VSD ASD PDA VSD/ASD VSD/PDA PPH APW TAPVC Total
35 27 3 4 1 12 1 1 84
Abbreviations as in Table I.
Table III. High-velocity tricuspid regurgitation: Moderate to severe Diagnosis
No.
VSD ASD PDA VSD/ASD VSD/PDA PPH APW TAPVC Total
32 24 4 5 3 12 0 0 80
Abbreviations as in Table I.
the incompetent pulmonary and tricuspid valves resided in the right side of the heart. The patients were reevaluated in our clinic by the coauthors at intervals of 6 months or less. Follow-up ranged from 2 years to 17 years (average 8.8 years). The observation period comprised approximately 1640 patient-years. Patients
who were not reevaluated by us within 1 year (generally those residing outside our area) were contacted, as were their physicians. The patients and physicians responded to a questionnaire designed to disclose the lowest probability of infective endocarditis. The questionnaire was modeled after the questionnaire used in the second Natural History Study of Congenital Heart Defects.7 Fourteen patients were lost to follow-up after observation periods of 5 to 11 years. Forty-one patients (24%) died during the course of observation. Necropsies were performed in 17 patients by the same cardiac pathologist. Two-dimensional transthoracic echocardiography with color flow imaging and spectral Doppler interrogation were performed in all patients, usually on several occasions. Twelve patients had transesophageal echocardiograms for reasons other than suspected infective endocarditis. Echocardiography established or confirmed the diagnosis of the basic congenital cardiac malformation and excluded high-velocity turbulent flow at sites other than the pulmonary or tricuspid valve. The pulmonary and tricuspid valves were examined for morphologic abnormalities by echocardiography in all cases and on gross examination in the 17 necropsy cases. The presence and degree of pulmonary and tricuspid regurgitation were determined by color flow imaging by the use of established criteria.8-10 For the purpose of this study, the degree of pulmonary and tricuspid regurgitant flow was characterized as absent or mild and moderate to severe. Peak velocity of regurgitant flow was based on color flow–guided, continuous wave Doppler echocardiography.8-11 Pulmonary arterial diastolic pressure and right ventricular systolic pressure were elevated by definition because of pulmonary vascular disease. The minimum pulmonary regurgitant end-diastolic velocity of 3.2 m/sec and the minimum tricuspid regurgitant peak systolic velocity of 4.7 m/sec established that regurgitant flow velocity was well above normal.8-11 In addition to the echocardiographic criteria, Graham Steele murmurs of pulmonary hypertensive pulmonary regurgitation and high-frequency holosystolic murmurs of high-pressure tricuspid regurgitation were routinely sought. Coexisting aortic regurgitation was excluded by Doppler echocardiography with color flow imaging. Criteria for the clinical diagnosis of infective endocarditis included an appropriate febrile illness, appropriate physical signs, positive blood cultures,1,3,4,12,13 and convincing evidence of a vegetation on transthoracic or transesophageal echocardiography.14,15 Analogous data were assembled on the 1174 UCLA Adult Congenital Heart Disease Registry patients not included in the above formal study. Forty-five percent were women and 55% were men. Ages ranged from 18 years to 95 years (mean ± SD, 42 ± 11 years). Follow-up ranged from 1 year to 20 years (mean ± SD, 11.1 ± 3.9 years). The observation period comprised approximately 11,860 patient-years. Seventy-two patients were lost to follow-up or died. The infective endo-
American Heart Journal Volume 136, Number 1
Figure 1
Necropsy specimen from 38-year-old woman with nonrestrictive ventricular septal defect and Eisenmenger reaction. Arrows point to nodular thickening along high-pressure closure line of inherently normal tricuspid valve (TV).
carditis questionnaire used for the pulmonary vascular disease study group (see earlier) was not used. The relative proportion of patients developing infective endocarditis and the number of episodes per 1000 patientyears were determined for the two patient groups, that is, the pulmonary vascular disease study group and the remaining patients in the registry.16-19 Exact 95% confidence intervals (95% CI) were calculated, and the significance of rate ratios was determined by using EpiInfo computer software (version 6.02, Centers for Disease Control, Atlanta, Ga.). Probability values are two tailed.
Results The tricuspid valves had inherently normal structure on the basis of echocardiography in all of the 186 study patients and on the basis of necropsy examination in 17 patients. Fifteen of the 17 necropsy specimens had nodular changes along the tricuspid leaflet closure lines that were considered secondary responses to chronic high-pressure tricuspid valve closure (Fig. 1). Despite high-velocity turbulent regurgitant flow, a jet impact site could not be identified in the 17 necropsy specimens. Tricuspid regurgitation was mild or absent in 106 patients (Table I) and moderate to severe in 80 (Table III). Tricuspid regurgitant peak systolic velocity ranged from 4.7 to 5.8 m/sec. No patient had documented infective endocarditis irre-
Dodo et al. 111
Figure 2
Necropsy specimen from 54-year-old woman with nonrestrictive ventricular septal defect (VSD) and Eisenmenger reaction. Paired bold arrows point to nodular thickening along high-pressure closure line of inherently normal pulmonary valve (PV).
spective of the presence or degree of high-velocity turbulent regurgitant flow. The pulmonary valves had inherently normal structure on the basis of echocardiography in all 186 patients and on the basis of necropsy examination in 17 patients. Despite high-pressure closure, only 1 of the 17 pulmonary valves examined at necropsy had secondary changes (Fig. 2), and despite high-velocity turbulent regurgitant flow, a jet impact site could not be identified in the 17 necropsy specimens. Pulmonary regurgitation was mild to absent in 102 patients (Table I) and moderate to severe in 84 (Table II). Pulmonary regurgitant end-diastolic velocities ranged from 3.2 to 3.8 m/sec. Infective endocarditis was identified in 1 patient, a 33-year-old cyanotic woman with pulmonary vascular disease and a nonrestrictive ostium secundum atrial septal defect, who developed Staphylococcus aureus endocarditis on the pulmonary valve while actively engaged in intravenous drug abuse. Transthoracic echocardiography identified a mobile vegetation attached to a moderately incompetent pulmonary valve. In a previous echocardiogram, the pulmonary valve was considered structurally normal and mildly incompetent.
112 Dodo et al.
Assuming that pulmonary valve infective endocarditis in the one patient was due to risk incurred by pulmonary vascular disease rather than to risk incurred by intravenous drug abuse on an inherently normal pulmonary valve, the rate of infective endocarditis in the primary study group was 0.54% (1 of 186) of patients or 0.61 episodes (1 of 640) per 1000 patient-years (95% CI 0.0 to 3.39). In the 1174 patients with congenital heart disease who were not included in the study group, 76 patients experienced 85 episodes of infective endocarditis. One patient experienced five episodes, and five patients experienced two episodes each. Thus the rate of infective endocarditis was 6.47% (76 of 1174) of patients or 7.17 episodes (85 of 11,860) per 1000 patient-years. On the basis of these estimates, the cumulative rate of infective endocarditis among the 1174 patients was almost 12 times greater than the rate among the study patients with pulmonary vascular disease (rate ratio 11.75; 95% CI 1.64 to 84.35; p = 0.002). Among the 76 patients who experienced infective endocarditis, 35 were unoperated (46%) and 41 were postoperative (54%). In the 35 unoperated patients, the diagnostic categories that served as the substrates for infective endocarditis included in relative order of frequency: tetralogy of Fallot, 24% (8% with pulmonary stenosis, 16% with pulmonary atresia); doubleoutlet right ventricle with pulmonary stenosis, 12%; restrictive ventricular septal defect, 12%; tricuspid atresia with normally related great arteries and subpulmonary stenosis, 10%; bicuspid aortic valve, 10%; congenitally corrected transposition of the great arteries, 8% (5% with pulmonary stenosis, 3% with pulmonary atresia); univentricular heart (left ventricular morphology) with pulmonary stenosis, 5%; and all other categories single examples, 21%. Of the 41 postoperative patients who had infective endocarditis, 22 had undergone palliative procedures (systemic to pulmonary arterial shunts, especially Blalock-Taussig). Nineteen of the 41 postoperative patients had reparative surgery for complex cyanotic congenital heart disease leaving behind residua and sequelae, including conduits and prosthetic valves.
Discussion High-velocity turbulent flow is regarded as a major predisposing cause of infective endocarditis either at the site generating the high-velocity turbulence (increased rate of shear) or at a jet impact site.1-4 Susceptibility at either site is believed to depend on endothelial damage, with risk determined by the
American Heart Journal July 1998
degree of damage to endothelial surfaces and by the pathogenicity (virulence) of the microorganism.3,4 The degree of endothelial damage assumes more importance when the microorganism is of low virulence and the bacteremia is transient and assumes less importance when the microorganism is of high virulence and the bacteremia is sustained or recurrent.3 Aortic regurgitation and mitral regurgitation are considered risk substrates for infective endocarditis because of high-velocity turbulent regurgitant flow.1,3,4 High-pressure pulmonary regurgitation (high-velocity turbulent diastolic flow) or high-pressure tricuspid regurgitation (high-velocity turbulent systolic flow) might be regarded as risk substrates analogous to aortic and mitral regurgitation. Inherently normal pulmonary valves are often rendered incompetent by pulmonary hypertension, but inherently normal trileaflet aortic valves typically remain competent even in the presence of substantial systemic hypertension. Such aortic valves have not been considered susceptible to infective endocarditis. However, in tetralogy of Fallot, especially with pulmonary atresia, initially competent trileaflet aortic valves housed in an enlarged aortic root accrue aortic regurgitation decade by decade of adult life.20 Among the 76 nonstudy patients who experienced infective endocarditis, 24% had tetralogy of Fallot, 8% with pulmonary stenosis and 6% with pulmonary atresia. The incidence of aortic regurgitation was highest in the latter category. Interestingly, inherently normal mitral valves are frequently rendered incompetent because left ventricular dilatation (heart failure or dilated cardiomyopathy, for example) results in lateral migration of the papillary muscles and because depressed systolic function prevents adequate reduction in mitral annular circumference.21 Infective endocarditis on these incompetent mitral valves is not known to incur insofar as we could determine, although susceptibility has not been systematically examined. Apart from intravenous drug abuse associated with sustained, recurrent bacteremia with high-virulence S. aureus, infective endocarditis rarely involves cardiac valves that are normal in structure and function.3,12,22 There is a unique propensity for intravenous drug abuse infective endocarditis to involve normal tricuspid and pulmonary valves, although normal left-sided valves are by no means spared.22,23 A pivotal question in our study is whether the single patient who incurred S. aureus infective endocarditis during active intravenous drug abuse was an example of infection
American Heart Journal Volume 136, Number 1
on a structurally and functionally normal pulmonary valve or whether pulmonary hypertension contributed to or caused the infection by lowering the threshold to sustained bacteremia with a high-virulence organism. Importantly, a first episode of infectious endocarditis has a far grater propensity to involve the tricuspid valve (44%) than the pulmonary valve (3%).22 If it is assumed that the single example of infective endocarditis in our study group was related to chronic high-pressure closure of a pulmonary valve that was judged by echocardiography to have been structurally normal and mildly incompetent before infection, then the cumulative rate of infective endocarditis was 0.061% or 0.61 per 1000 patient-years (95% CI 0.0 to 3.39). Conversely, if it is assumed that this single case was an example of right-sided infective endocarditis unrelated to pulmonary hypertension, then the rate of infection was 0 per 1640 patient-years in our study population (95% CI 0.0 to 2.25). Prolonged high-pressure closure associated with pulmonary hypertension can induce morphologic changes on tricuspid or pulmonary valves of inherently normal structure. In our necropsy patients, minor thickening of the tricuspid valve closure line was commonly so related (Fig. 1), but analogous changes were found on only one pulmonary valve (Fig. 2) despite the probability of inherent pulmonary cuspal inequality. Assuming the presence of these structural changes, infective endocarditis did not occur in any of the 186 patients apart from the single case of intravenous drug abuse irrespective of the presence or degree of high-velocity tricuspid or pulmonary regurgitation. Nor could we find in the literature an example of infective endocarditis involving tricuspid or pulmonary valves that were inherently normal but exposed to high-velocity pulmonary hypertensive regurgitant flow. On the basis of our necropsy observations, highvelocity regurgitation across tricuspid or pulmonary valves of inherently normal structure apparently causes little or no increase in shear rate at valve level and little or no jet impact on endothelial surfaces remote from these valves. Accordingly, endothelial damage—a sine qua non of susceptibility3,4—may be insignificant or absent. When considering rare events such as the incidence of infective endocarditis, risk can be estimated by comparing cases (herein the pulmonary vascular disease study patients) per 1000 patient-years with the cases of infective endocarditis per 1000 patient-years in the general population.16-19 To accumulate sufficient
Dodo et al. 113
patient-years to validate such a study would require a multicenter collaboration. However, we should not lose sight of the fact that there is a difference between years of observation derived from large numbers of patients observed for short periods of time versus smaller numbers observed for long periods of time, as in our study. Alternatively, infective endocarditis risk might be estimated by comparing our pulmonary vascular disease study patients (186 of 1360) with the additional patients (1174 of 1360) in the UCLA Adult Congenital Heart Disease Registry. Because a substantial portion of the 1174 patients had been instructed regarding prophylaxis for infective endocarditis, and because none of the pulmonary vascular disease study patients has been so advised, a comparison would provide the maximal rather than the minimal estimate of relative risk.
Limitations of the study and clinical implications regarding prophylaxis When considering rare events such the occurrence of infective endocarditis, the small size of our study is a limitation, although the long duration of follow-up is a strength. To resolve questions of relative risks that lie in the range of 0 to 1 event per 1000 patient-years would require a multicenter study involving large numbers of patients. We chose instead to report our clinical/epidemiologic observations without reaching a final conclusion regarding risk of infective endocarditis on inherently normal tricuspid and pulmonary valves rendered incompetent by pulmonary hypertension. During approximately 1640 patient-years of observation, we identified 1 case of infective endocarditis on the pulmonary valve of a habitual intravenous drug abuser. If this case is included, the cumulative rate of infective endocarditis was 0.061% or 0.61 per 1000 patient-years (95% CI 0.0 to 3.39 per 1000 patientyears). In the general adult population of the United States, the annual incidence of infective endocarditis is reportedly 0.005%, or 1 case in 20,000 persons per year.16-18 In an urban European population, the annual incidence is reportedly 0.006%, or 1.2 cases in 20,000 persons per year.19 Analysis of the 1174 additional Adult Congenital Heart Disease Registry patients who were apart from those in our study population disclosed 85 episodes of infective endocarditis among 76 patients during an observation period of approximately 11,860 patientyears. Given the rate of infective endocarditis in the 186 pulmonary vascular disease study patients versus
American Heart Journal July 1998
114 Dodo et al.
the rate in the 1174 additional registry patients, the rate was 10.5 to 12 times higher in the latter group. Our clinical and epidemiologic observations leave open the question of whether high-velocity turbulent flow across inherently normal tricuspid or pulmonary valves rendered incompetent by pulmonary hypertension incurs a risk of infective endocarditis. These substrates may be low risk or no risk.24 However, until this clinically important question is satisfactorily resolved, prophylaxis against infective endocarditis remains advisable.
References 1. Dajani AS, Taubert KA, Wilson W, Bolger AF, Bayer A, Ferrieri P, et al. Prevention of bacterial endocarditis: recommendations by the American Heart Association. Circulation 1997;96:358-66. 2. Rodbard S. Blood velocity and endocarditis. Circulation 1963;27:18-28. 3. Steckelberg JM, Wilson WR. Risk factors for infective endocarditis. Infect Dis Clin North Am 1993;7:9-19. 4. Child JS, Perloff JK, Kubak B. Infective endocarditis: risks and prophylaxis. In: Perloff JK, Child JS. Congenital heart disease in adults. 2nd ed. Philadelphia: WB Saunders Co; 1998. p. 129-43. 5. Gersony WM, Hayes CJ, Driscoll DJ, Keane JF, Kidd L, et al. Bacterial endocarditis in patients with aortic stenosis, pulmonary stenosis, or ventricular septal defect. Circulation 1993;87(suppl I):I-121-6. 6. Corone P, Levy A, Hallali P, Davido A, Wyler Y, Corone A. A propos de 54 cas d’endocardites infectieuses observès en 32 ans sur une population de 2038 cardiopathies congènitales. Arch Mal Coeur 1989;82:779-84. 7. O’Fallon WM, Crowson CS, Rings LJ, Weidman WH, Driscoll DJ, et al. Second natural history study of congenital heart defects: materials and methods. Circulation 1993;87(suppl I):I-4-15. 8. Reeder GS, Currie PJ, Hagler DJ, Tajik AJ, Seward JB. Use of Doppler techniques (continuous-wave, pulse-wave, and color flow imaging) in the noninvasive hemodynamic assessment of congenital heart disease. Mayo Clin Proc 1986;61:725-44. 9. Sahn DJ. Application of color flow mapping in pediatric cardiology. Cardiol Clin 1989;7:255-64. 10. Otto CM, Pearlman AS. Valvular regurgitation: diagnoses, quan-
11. 12.
13. 14.
15.
16.
17.
18.
19.
20. 21. 22.
23. 24.
tification and clinical approach. In: Textbook of clinical echocardiography. Philadelphia: W.B. Saunders Company; 1995. p. 243-78. Schiller NB. Pulmonary artery pressure estimation by Doppler and two-dimensional echocardiography. Cardiol Clin 1990;8:277-87. Kubak BM, Nimmagadda AP, Holt CD. Advances in medical and antibiotic management of infective endocarditis. Cardiol Clin 1996; 14:405-36. Durack DT. Prevention of infective endocarditis. N Engl J Med 1995;332:38-44. Shively BK, Gurule FT, Roldan GA, Leggett JH, Schiller NB. Diagnostic value of transesophageal compared with transthoracic echocardiography in infective endocarditis. J Am Coll Cardiol 1991;18:391-7. Sanfelippo AJ, Picard MH, Newell JB, Rosas E, Davidoff R, Thomas JD, et al. Echocardiographic assessment of patients with infectious endocarditis: prediction of risk for complications. J Am Coll Cardiol 1991;18:1191-9. MacMahon SW, Roberts JK, Kramer-Fox R, Zuker DM, Roberts RB, Deuereux RB. Mitral valve prolapse and infective endocarditis. Am Heart J 1987;113:1291-8. Kozak LJ, Moien M. Detailed diagnoses and procedures for patients discharged from short-stay hospitals: United States, 1983. Hyattsville, Md.: National Center for Health Statistics; 1985, DHHS publication No. (PHS) 85-1643. (Vital and Health Statistics). Lawrence L. Detailed diagnoses and procedures for patients discharged from short-stay hospitals: United States, 1984. Hyattsville, Md.: National Center for Health Statistics; 1985, DHHS publication No. (PHS) 86-1647. (Vital and Health Statistics). Hogevik H, Alaison L, Anderson R, Lindberg J, Alestig K. Epidemiologic aspects of infective endocarditis in an urban population: a 5year prospective study. Medicine (Baltimore) 1995;74:324-39. Marelli AJ, Perloff JK, Child JS, Laks H. Pulmonary atresia with ventricular septal defect. Circulation 1994;89:243-51. Perloff JK, Roberts WC. The mitral apparatus: functional anatomy of mitral regurgitation. Circulation 1972;46:227-39. Levine DP, Crane LR, Zervos MJ. Bacteremia in narcotic addicts at the Detroit Medical Center, II: infectious endocarditis: a prospective comparative study. Rev Infect Dis 1986;8:374-96. Dressler FA, Roberts WC. Infective endocarditis in opiate addicts: analysis of 80 cases studied at necropsy. Am J Cardiol 1989;63:1240-57. Michael PL, Acar J. Native cardiac disease predisposing to infective endocarditis. Eur Heart J 1995;16:2-6.