echocardiographic screening for rheumatic heart disease: age matters
TRANSCRIPT
International Journal of Cardiology 168 (2013) 888–891
Contents lists available at ScienceDirect
International Journal of Cardiology
j ourna l homepage: www.e lsev ie r .com/ locate / i j ca rd
Echocardiographic screening for rheumatic heart disease: Age matters
Abdoul Kane a,1, Mariana Mirabel b,c,d,⁎,1, Kamadore Touré a, Marie-Cécile Périer b,c, Samia Fazaa e,Muriel Tafflet b,c, Nicole Karam b,c,f, Ibrahim Zourak a, Dior Diagne a, Alassane Mbaye a, Moussa Kane a,Bouna Diack a, Xavier Jouven b,c, f, Eloi Marijon b,c, f
a Hôpital Général Grand Yoff, Dakar, Senegalb Paris Cardiovascular Research Center, Inserm U970, European Georges Pompidou Hospital, Paris, Francec Paris Descartes University, Paris, Franced University College London, London, United Kingdome Centre Hospitalo-Universitaire Mongi Slim, La Marsa, Tunisiaf Cardiology Department, European Georges Pompidou Hospital, Paris, France
⁎ Corresponding author at: Paris Cardiovascular ResearEuropéen Georges Pompidou, 56 rue Leblanc, 75737 Pari95 97 02; fax: +33 1 53 98 79 54.
E-mail address: [email protected] (M. Mira1 These authors equally contributed to this work.
0167-5273/$ – see front matter © 2012 Elsevier Irelandhttp://dx.doi.org/10.1016/j.ijcard.2012.10.090
a b s t r a c t
a r t i c l e i n f oArticle history:
Received 15 July 2012Received in revised form 27 September 2012Accepted 28 October 2012Available online 13 December 2012Keywords:Acute rheumatic feverUltrasoundStreptococcusHeart valve diseaseAdolescent
Background: Echocardiography is emerging as a screening tool for rheumatic heart disease (RHD) in endemicregions. The vast majority of surveys have been limited to children. We ought to appreciate the interest ofincluding adolescents in their late teens in such school screening programmes.Methods: School-based echocardiography cross-sectional survey conducted in Dakar, Senegal (March 2010).A total of 2004 school attendees were randomly selected and enrolled in the study, among which 1116 wereaged 5–15 years old (group 1), and 888 were 16–18 years old (group 2). Case detection rates and phenotypeof RHD were compared according to age groups.Results: A total of 22 youngsters were suspected by on-site echocardiography, 12 in group 1 and 10 in group2. Among the 12 RHD cases suspected on-site in group 1, 6 (50%) were eventually considered as confirmedRHD, compared to 9 out of 10 (90%) in group 2, giving prevalence rates of 5.4 (CI 95% 2.0–11.7) and 10.1(CI 95% 4.6–19.2) per 1000 in group 1 and group 2, respectively. The proportion of marked/advanced lesions
was 33% in group 1, and 89% in group 2 (p=0.08). Mean concordance rates between the 3 reviewers were40% for group 1, compared to 93% in group 2 (p=0.05).Conclusions: Extension of screening to adolescents in their late teens should be considered with interest inthe light of the higher prevalence of the disease and relative clarity of subclinical cardiac lesions that couldbe more easily detected in the field.© 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Rheumatic heart disease (RHD) remains a major health problemin developing countries, affecting the poorest regions, namelysub-Saharan Africa [1–3]. Pathogenesis of the disease involves expo-sure to group A streptococcus (GAS), abnormal inflammatory response,and susceptibility resulting in the damage of the cardiac valves. RHD isresponsible for heart disease in the young, including heart failure andinfective endocarditis, and presents a significant burden in women ofchildbearing age, leading to approximately 250,000 premature deathsper year worldwide [4,5].
Whilst themajority of developing nations consume a vast amount ofresources in treating most severe cases that require intervention [6],
ch Center, Inserm U970, Hôpitals CEDEX 15, France. Tel.: +687
bel).
Ltd. All rights reserved.
prevention policies have proven cost-effective, especially secondaryprophylaxis within comprehensive programmes [7–9]. To date, activesurveillance within efficient register-based control programmes shouldbe one of the major players in controlling the disease [10]. Early detec-tion is paramount to implement secondary prevention and limit theevolution towards severe valve damage [11]. In this context, echocardi-ography has emerged as a potential tool in high-burden populations[12,13], allowing the detection of subclinical disease as well as the con-firmation of clinically audible lesions. RHD prevalence is known to sig-nificantly increase with age because of the cumulated effect ofrepeated exposure to GAS [14].
Most recent screening programmes have focused on young schoolchildren up to 14 or 15 years old [15–19]. To date only two studies haveincluded a very small proportion of adolescents over 15 [12,20]. We hy-pothesized that extending the survey to older teenagers (16–18 years)may be particularly attractive, by increasing the case-detection ratesand by easily identifying valve lesions on-site. We therefore conducted aschool-based echocardiography cross-sectional survey in Dakar, Senegal,through a large randomized sample from the school community, and
Table 1Demographic characteristics and phenotype of RHD according to age (groups 1 and 2).Group 1: 5–15 years old; Group 2, 16–18 years old.
Group 1n=1116
Group 2n=888
Age, mean (SD) 11.6 (2.0) 17.1 (1.0)Gender, male, n (%) 622 (55.7) 412 (46.4)Prevalences
▪Prevalence, on-site, per 1000 (CI 95%) 10.8 (4.7–16.9) 11.3 (4.3–18.3)▪Confirmed prevalence, n per 1000 (CI 95%)⁎ 5.4 (2.0–11.7) 10.1 (4.6–19.2)
Proportion of marked lesions, n (per 1000)⁎⁎ 2 (1.8) 8 (9.0)Mean concordance rates (%)⁎⁎⁎ 40 93
⁎ Defined after review of the confirmation hospital-based scan and agreement of atleast 2 out of 3 reviewers.⁎⁎ Defined as moderate or severe left-sided mitral or aortic valve disease, and/ormultivalvular minimal involvement (i.e., the combination of mitral and aortic valvechanges deemed significant), and/or significant mitral stenosis (valve areab2.5 cm2
by planimetry on parasternal short axis view).⁎⁎⁎ Mean concordance rates between the 3 reviewers (R1/R2, R2/R3 and R1/R3) werecalculated for each group.
2,012 randomlyselected among6,543 school-
children
2,004 underwentscreening
Group 1
1,116 children aged5-15 years
12 cases suspectedon site
(2 cases with amurmur)
6 cases confirmedafter formal hospital
review
Group 2
888 children aged16-18 years
10 cases suspectedon site
(1 case with amurmur)
9 cases confirmedafter formal hospital
review
8 declined
Fig. 1. Flow chart.
889A. Kane et al. / International Journal of Cardiology 168 (2013) 888–891
analysed data according to two groups of different ages (5–15 and16–18 years).
2. Methods
2.1. Setting and population
We conducted a survey from March 1st to March 31st 2010 in schools from urbanand suburban Dakar (2,452,656 inhabitants according to the 2005's census) after ap-proval of the Ethics Committee of the Academy Inspection under the Ministry of Edu-cation [21]. Using Epi Info software (version 3.2.2), 4 public schools among primary andmiddle schools of Dakar and 2 high schools were randomly selected. Only children withinformed parental or guardian consent were offered to take part in the study. A total of2004 children were finally enrolled, among which 1116 were aged 5–15 years (group1), and 888 were 16–18 years old (group 2).
2.2. Aims of the study
The aim of the study was to evaluate to what degree extending the survey to apopulation in its late teens (16–18 years old) would increment case-detection ratesand modify the presentation of subclinical valve lesions.
2.3. Clinical and echocardiography protocol
All children underwent clinical and echocardiographic examinations at school in aspecifically arranged room by the same team (IZ, DD, and AM under the supervision ofAK). Echocardiography was performed on-site with portable equipment (SonosSonosite MicroMax® using a 3 MHz transducer). When RHD was suspected on-site, aconfirmation scan was carried out at the hospital using a non-portable machine(Imagic Agile Kontron®). The confirmation study was recorded for further review by3 independent experts (OD, ML, and ML) with a positive confirmed RHD case whenat least 2 reviewers agreed. Secondary prophylaxis by benzathine penicillin G wasstarted in all positive cases [22].
2.4. Echocardiographic criteria
Definition of echocardiographic features of RHD included the combination ofDoppler criteria and at least two mitral valve morphological changes. Doppler criteriawere defined as follows: mitral or aortic valve regurgitation was considered significantif seen in at least two planes on colour Doppler, excluding closure jets. Qualitative mor-phological criteria of the mitral valve included: (i) significant thickening of the valveleaflets either at the tips or through their bodies; (ii) thickening and/or shortening ofthe mitral subvalvular apparatus; and (iii) mitral valve motion abnormalities as leafletrestriction. Mitral stenosis (MS) was defined as a valve areab2.5 cm2 by planimetry onthe parasternal short axis view. Quantification of left-sided valve disease was under-taken according to current guidelines [23].
Marked RHD (either clinical or subclinical) was defined as moderate or severe mi-tral or aortic valve disease, and/or multivalvular minimal involvement (i.e., the combi-nation of mitral and aortic valve changes deemed significant), and/or significant valvestenosis (mitral valve areab2.5 cm2 by planimetry on parasternal short axis view).
2.5. Statistical analysis
All data were entered and analysed using SPSS version 16. Descriptive variables arepresented as means±SD; sample means were compared with an independent-sampleStudent's t-test. Regarding subclinical cases, mean concordance rates between the 3 re-viewers (R1/R2, R2/R3 and R1/R3) were calculated for each group. The exact Fisher testwas used to compare the frequencies. Calculation of 95% exact confidence intervals forprevalence was performed. A two-sided p value of less than 0.05 was considered to in-dicate statistical significance. The authors of this manuscript have all certified that theycomply with the principles of ethical publishing.
3. Results
Subjects' characteristics, according to groups, are depicted in Table 1,and flow chart of the study in Fig. 1. The mean age was 11.6±2 ingroup 1, and 17.1±1 in group 2, with a sex ratio (M/F) of 1.3 and0.9 in groups 1 and 2, respectively.
Twenty-four children (1.2% of the overall population) presentedwith a murmur thought to be pathological. Among these, 6 cases(25%) were related to organic lesions on echocardiography: 3 toRHD (2 children in group 1, and 1 adolescent in group 2), and 3 tocongenital heart disease. In the remaining 18 children, no cardiac ab-normality was confirmed on echocardiography.
Echocardiographic screening findings according to age groups aresummarized in Table 1 and schematized in Fig. 2. In group 1, 12
suspected RHD cases were detected on-site using portable echocardi-ography. Six out of these cases (50%) were eventually confirmed aftertraditional echocardiography and review process giving an overallRHD prevalence of 5.4 (CI 95% 2.0–11.7) per 1000. In group 2, 10 chil-dren with suspected RHD were detected on-site using portable
0
2
4
6
8
10
12
14
16
18
20
Group 1 (6-15 year) Group 2 (16-18 year)
Pre
vale
nce
of
RH
D, P
er 1
,000
On-Site Suspected RHD
Confirmed RHD
Fig. 2. Prevalence rates according to age group, and ability of screening using portabledevice in the field to detect confirmed RHD cases.
890 A. Kane et al. / International Journal of Cardiology 168 (2013) 888–891
echocardiography. Nine out of these cases (90%) were subsequentlyconfirmed after traditional echocardiography and review process giv-ing an overall RHD prevalence of 10.1 (CI95% 4.6–19.2) per 1000.
Regarding the proportion of marked RHD, among the 6 confirmedRHD cases in group 1, only 2 (33%) presented such features. By con-trast, in group 2, among the 9 confirmed cases of RHD, 8 (89%)presented marked disease (p=0.08), of which 3 with some degreeof mitral stenosis.
Regarding the accuracy to eventually detect subclinical RHD cases,we analysed the data from the 3 reviewers. During the review processof subclinical suspected cases in group 1 (10 cases), the mean concor-dance rates between reviewers were 60%, 40% and 20%, to finallyidentify 4 confirmed RHD subclinical cases. In group 2 (8 cases), theconcordance rates between the 3 reviewers were 100%, 90% and90%, to finally identify 7 confirmed RHD subclinical cases.
4. Discussion
Our results suggest the particular interest of considering adoles-cents in their late teens for RHD screening in endemic regions, sinceprevalence in this group appears to be higher with more pronouncedvalve lesions when compared to their younger counterpart, thereforeeasily identified in the field by means of portable echocardiographyalone (without need for further confirmation in hospital settings).
The trend towards a higher prevalence of subclinical RHD in theolder group is in agreement with another study [24]. The prevalenceof symptomatic disease is known to increase with age, peaking inyoung and middle-aged adults [25,26]. This is in keeping with repeat-ed GAS exposure, which leads to insidious worsening of valve damagethrough the years.
The presentation of RHD varies with age, mitral regurgitationbeing the most common valve disease in the young, with increasingproportion of mitral stenosis and aortic valve involvement in adult-hood [14,27]. Our data also suggested that adolescents in their lateteens mainly presented with marked rheumatic lesions of the valves,including 3 cases of juvenile mitral stenosis and combined aortic andmitral valve involvement. Consistently, the concordance rates be-tween readers were particularly high in the older group. This is ofmajor importance since it has practical implications as to futurelarge echocardiography-based screening surveys. Indeed, the factthat most adolescents in their late teens may not need a confirmationscan appears particularly attractive in remote locations where techni-cal expertise and access to health care facilities are lacking [28].
The overall prevalence of RHD observed in Dakar seemed somewhatlower in our survey than in other sub-Saharan countries [12,20]. Theremay be a true difference in the burden of the disease. For example, Sen-egal presents higher development standards than Mozambique as forGDP per capita, and this is reinforced by the lower clinical RHD preva-lence found in Senegal as well (approximately 1.5 per 1000). Interpre-tation of echocardiography criteria is not always simple and may beparticularly subjective when lesions are mild. The proportion of verymild lesions may have been underestimated in our survey when com-pared to others due to the fact that reviewers were not particularlytrained to detect subclinical RHD, and probably more likely to considermore advanced subclinical cases. This was illustrated by the low level ofagreement between readers in the younger group.
A diagnostic test should ideally be simple and reproducible to beapplied in poorly resourced settings [28]. Although standardized,the use of complex echocardiography criteria requires sophisticatedequipment and expertise, both lacking in most regions where RHDis endemic [29]. In our study, most cases suspected on-site were sub-sequently confirmed in adolescents in their late teens. Although someauthors advocate a two-step screening procedure for clinical pur-poses [20], portable echocardiography alonemay be considered an at-tractive solution in groups with more marked valvular lesions. Thespecificity may be diminished by this approach but it would be appli-cable in remote locations and particularly suitable to the vast majorityof the target population in sub-Saharan countries.
Finally, our results highlight the importance of a broader age-range for RHD screening, and invite to re-think how RHD surveillanceshould be conducted. Most echo-screening surveys included childrenup to 14 or 15 years of age [16–20,30]. The design of such studies wasprobably influenced by age-distribution of ARF, and the lower schoolattendance in adolescents in their late teens [25].
Our study presents a number of limitations. The echocardiographycriteria are similar to that used in other studies but do not meet theWorld Heart Federation's standards, the latter has been agreed uponafter the completion of the survey [12,29]. Our prevalence resultsdid not meet statistical significance, likely due to a lack of power.Although the management of subclinical RHD is subject to debate,we decided to offer secondary prophylaxis to all confirmed cases [31].
5. Conclusion
More than reemphasizing the higher sensitivity of ultrasounds indetecting valvular lesions, our results suggest the importance of con-sidering adolescents in their late teens in RHD screening strategiessince they present with a higher prevalence and particularly markedvalve lesions, likely to be detected in the field.
Acknowledgements
We would like to thank Ousmane Dièye, Mohamed Leye, andMariame Ly for reviewing the confirmation scans.
References
[1] Essop MR, Nkomo VT. Rheumatic and nonrheumatic valvular heart disease: epi-demiology, management, and prevention in Africa. Circulation 2005;112(23):3584–91.
[2] Marijon E, Mirabel M, Celermajer DS, Jouven X. Rheumatic heart disease. Lancet2012;379(9819):953–64.
[3] Longo-Mbenza B, Bayekula M, Ngiyulu R, et al. Survey of rheumatic heart diseasein school children of Kinshasa town. Int J Cardiol 1998;63(3):287–94.
[4] Carapetis JR, Steer AC, Mulholland EK, Weber M. The global burden of group Astreptococcal diseases. Lancet Infect Dis 2005;5(11):685–94.
[5] Diao M, Kane A, Ndiaye MB, et al. Pregnancy in women with heart disease insub-Saharan Africa. Arch Cardiovasc Dis 2011;104(6–7):370–4.
[6] Ferratini M, Marianeschi S, Santoro F, et al. Valvulopathies in sub-Saharan Africanchildren: patterns, humanitarian interventions and cardiac surgical problems. IntJ Cardiol May 10 2013;165(2):237–41.
[7] Arguedas A, Mohs E. Prevention of rheumatic fever in Costa Rica. J Pediatr1992;121(4):569–72.
891A. Kane et al. / International Journal of Cardiology 168 (2013) 888–891
[8] Nordet P, Lopez R, Duenas A, Sarmiento L. Prevention and control of rheumaticfever and rheumatic heart disease: the Cuban experience (1986–1996–2002).Cardiovasc J Afr 2008;19(3):135–40.
[9] North DA, Heynes RA, Lennon DR, Neutze J. Analysis of costs of acute rheumaticfever and rheumatic heart disease in Auckland. N Z Med J 1993;106(964):400–3.
[10] Steer AC, Carapetis JR. Prevention and treatment of rheumatic heart disease in thedeveloping world. Nat Rev Cardiol 2009;6(11):689–98.
[11] Celermajer DS, Chow CK, Marijon E, Anstey N, Woo KS. Cardiovascular diseases inthe developing world— prevalences, patterns and the potential of early detection.J Am Coll Cardiol 2012;60(14):1207–16.
[12] Marijon E, Ou P, Celermajer DS, et al. Prevalence of rheumatic heart diseasedetected by echocardiographic screening. N Engl J Med 2007;357(5):470–6.
[13] Marijon E, Tafflet M, Jouven X. Time to use ultrasound and not stethoscopes forrheumatic heart disease screening. Nat Clin Pract Cardiovasc Med Jul 2008;5(7):E1–3.
[14] Sliwa K, Carrington M, Mayosi BM, Zigiriadis E, Mvungi R, Stewart S. Incidence andcharacteristics of newly diagnosed rheumatic heart disease in urban Africanadults: insights from the heart of Soweto study. Eur Heart J 2010;31(6):719–27.
[15] Steer AC, Kado J, Wilson N, et al. High prevalence of rheumatic heart disease byclinical and echocardiographic screening among children in Fiji. J Heart ValveDis 2009;18(3):327–35.
[16] Carapetis JR, Hardy M, Fakakovikaetau T, et al. Evaluation of a screening protocolusing auscultation and portable echocardiography to detect asymptomatic rheu-matic heart disease in Tongan schoolchildren. Nat Clin Pract Cardiovasc Med2008;5(7):411–7.
[17] Bhaya M, Panwar S, Beniwal R, Panwar RB. High prevalence of rheumatic heartdisease detected by echocardiography in school children. Echocardiography2010;27(4):448–53.
[18] Saxena A, Ramakrishnan S, Roy A, et al. Prevalence and outcome of subclinicalrheumatic heart disease in India: the RHEUMATIC (Rheumatic Heart EchoUtilisation and Monitoring Actuarial Trends in Indian Children) study. Heart2011;97(24):2018–22.
[19] Webb RH, Wilson NJ, Lennon DR, et al. Optimising echocardiographic screeningfor rheumatic heart disease in New Zealand: not all valve disease is rheumatic.Cardiol Young 2011;21(4):436–43.
[20] Beaton A, Okello E, Lwabi P, Mondo C, McCarter R, Sable C. Echocardiographyscreening for rheumatic heart disease in Ugandan schoolchildren. Circulation2012;125(25):3127–32.
[21] Situation Economique et Sociale du Sénégal, Ministère de l'Economie et des Fi-nances. République du Sénégal: Agence Nationale de la Statistique et de laDémographie; June 1st 2011.
[22] WHO. Rheumatic fever and rheumatic heart disease — report of a WHO expertconsultation. Geneva: WHO; 2001.
[23] Bonow RO, Carabello BA, Kanu C, et al. ACC/AHA 2006 guidelines for the manage-ment of patients with valvular heart disease: a report of the American College ofCardiology/American Heart Association Task Force on Practice Guidelines (writingcommittee to revise the 1998 Guidelines for the Management of Patients With Val-vular Heart Disease): developed in collaboration with the Society of CardiovascularAnesthesiologists: endorsed by the Society for Cardiovascular Angiography and In-terventions and the Society of Thoracic Surgeons. Circulation 2006;114(5):e84-231.
[24] Marijon E, Celermajer DS, Tafflet M, et al. Rheumatic heart disease screening byechocardiography: the inadequacy of World Health Organization criteria for opti-mizing the diagnosis of subclinical disease. Circulation 2009;120(8):663–8.
[25] ParnabyMG, Carapetis JR. Rheumatic fever in indigenousAustralian children. J PaediatrChild Health 2010;46(9):527–33.
[26] Rizvi SF, Khan MA, Kundi A, Marsh DR, Samad A, Pasha O. Status of rheumaticheart disease in rural Pakistan. Heart 2004;90(4):394–9.
[27] Marcus RH, Sareli P, Pocock WA, Barlow JB. The spectrum of severe rheumatic mi-tral valve disease in a developing country. Correlations among clinical presenta-tion, surgical pathologic findings, and hemodynamic sequelae. Ann Intern Med1994;120(3):177–83.
[28] Mirabel M, Celermajer DS, Ferreira B, et al. Screening for rheumatic heart disease:evaluation of a simplified echocardiography-based approach. Eur Heart JCardiovasc Imaging 2012;13(12):1024–9.
[29] Remenyi B, Wilson N, Steer A, et al. World Heart Federation criteria for echocar-diographic diagnosis of rheumatic heart disease—an evidence-based guideline.Nat Rev Cardiol 2012;9(5):297–309.
[30] Reeves BM, Kado J, Brook M. High prevalence of rheumatic heart disease in Fijidetected by echocardiography screening. J Paediatr Child Health 2011;47(7):473–8.
[31] Marijon E, Celermajer DS, Jouven X. Management of patients with subclinicalrheumatic heart disease. Int J Cardiol May 29, 2009;134(3):295–6.