Is Rheumatic Heart Disease Still Endemic in Tunisia? A Review of 14-year Period and Future Projections

Houda Ben Ayed^1*, Sourour Yaich¹, Mariem Ben Hmida¹, Maissa Ben Jemaa¹, Maroua Trigui², Jihene Jedidi¹, Raouf Karray¹, Habib Feki², Yosra Mejdoub¹, Mondher Kassis¹ & Jamel Damak¹

¹Community Health and Epidemiology Department, Hedi Chaker University Hospital, University of Sfax, Tunisia
²Preventive Medicine and Hygiene Department, Hedi Chaker University Hospital, University of Sfax, Tunisia

*Correspondence to: Dr. Houda Ben Ayed, Community Health and Epidemiology Department, Hedi Chaker University Hospital, University of Sfax, Tunisia.

Copyright © 2018 Dr. Houda Ben Ayed, et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abbreviations
RHD: Rheumatic Heart Disease; APC: Annual Percentage Change; LcrI: Lower credible interval, UCrI: Upper credible interval; HCUH: Hedi Chaker University Hospital; HCIR: Hospital Crude Incidence Rate; 95% CI: 95% Confidence Interval; SD: Standard deviation; PY: Person-year.

Introduction
Rheumatic heart disease (RHD) ranks among the leading causes of non-communicable diseases in low-income and middle-income countries and causes up to 250000 premature deaths every year [1]. It results from a primary infection in the upper respiratory tract or from a re-infection or a relapse associated with an inappropriate inflammatory response to the group A beta hemolytic Streptococci [2]. According to the WHO, 12 million people worldwide are affected by rheumatic fever and rheumatic heart disease, among whom 79% live in the developing countries, particularly those in the African continent [3]. It continues to be a major public health challenge in limited resources countries, accounting for a great cardiovascular-related loss of disability-adjusted life-years worldwide [4]. RHD affects the most productive part of the population, which may lead to undermine national productivity. Its epidemiology parallels most indicators of poverty, notably poor hygiene, household crowding, poor living conditions, social inequality and limited access to healthcare. In high-income countries, the burden of RHD decreased markedly during the 20th century, attributable to socioeconomic development and advances in medical management [5].

Recently, there has been an increasing interest in the burden of RHD, driven in part by the growing costs attributable to this disease and the utmost need to yield standards level in cardio-vascular health. Epidemiological surveillance is particularly interesting in endemic area, allowing health authorities to prioritize preventive and curative interventions accordingly. Tunisia is recognized as an endemic area of RHD. Otherwise, updated and reliable data of RHD incidence in South of Tunisia are weakly documented. Therefore, our study aimed to determine the epidemiological specificities of RHD and to assess its chronological trends over time.

Methods

Study Design and Settings
We retrospectively collected data from the regional morbidity register of Hedi Chaker University Hospital (HCUH), South of Tunisia, from January 1^st, 2003 to December 31^st, 2016. The register was implemented by the community health and epidemiology department since 1990 and received information periodically from all departments of HCUH. Patients were hospitalized essentially in cardiology department, which included intensive care unit and applied the highest cardiologic expertise and advanced diagnostic investigations to make definitive diagnoses and provide medical care.

Inclusion Criteria and Data Collection
We included all newly diagnosed patients with RHD, who were hospitalized at HCUH during the study period. Patients ultimately diagnosed RHD were encoded according to the International Classification of Diseases-10 (ICD-10) (I01, I05, I06, I07, I08, I09). Patients hospitalized for acute articular rheumatism without carditis or other complications coded in ICD-10 as I00. I02 were not included. We included only patients living in Sfax governorate.

During the study period, the definitions of RHD were unified. The database variables included patients’ socio-demographic characteristics and hospitalization characteristics such as length of hospital stay, the discharge status (home return, transfer to other departments or others hospitals and death). The crude hospital incidence rate (CHIR) of RHD was calculated based on Tunisian National Institute of Statistics data and was expressed as the number per 100000 person-year (PY).

Chronological Trends and Projected Incidence Rates
In order to analyze trends in RHD hospitalizations, the Joinpoint Regression Analysis program, version 4.6.0.0 was performed. Joinpoint fits a linear regression model to the data to detect periods with statistically distinct log-linear trends over time. A significance level of 0.05 was used for the permutation test, which determines the minimum number of “joinpoints” necessary to fit the data. The annual percent change (APC) within each segment was calculated with 95% confidence interval (95% CI).

To perform RHD projection for 2026, an age-period cohort model was performed assuming a Poisson regression model for the counts of the cases. We estimated the mean number of new cases of RHD in each year, with lower and upper credible intervals (LCrI, UCrI).

Statistical Analysis
Statistical analysis was performed using SPSS.20 software. The Kolmogorov-Smirnov test was used to assess the distribution of quantitative variables. The results of quantitative variables were presented as means ± Standard deviation (SD) or medians and interquartile range (IQR). Those of qualitative variables were presented as numbers and percentages. We used Spearman test (Rho coefficient, p) to study the correlation between two quantitative variables. A p-value of <0.05 was considered statistically significant.

Results

Patients’ Characteristics
During the 14-year study period, we registered 145166 hospitalizations of any age, all diseases combined, among whom 1063 patients hospitalized for RHD (0.73%). We noted 635 women with RHD (59.7%), with a sex ratio of 0.7. The mean age of patients with RHD was 46 ± 17.8 years. There were 432 cases (40.6%) aged between 40 and 59 years (Table 1). Cardiology department hosted 1057 patients with RHD (99.4%). Home return was the most common discharge status (991 cases; 93.3%). The median hospital length of stay was 7 days (IQR, 4-13 days). There were 18 deaths, accounting for a case-fatality rate of 1.6%. Rheumatic mitral valve was the dominant RHD localization (547 cases; 51.5%) (Table 2).

N: Number; %: Percentage among all rheumatic heart disease hospitalizations; HCIR: Hospital crude incidence rate, HCMR: Hospital crude mortality rate; PY: person-year.

Incidence and Mortality Rates of RHD
The mean annual number of RHD new cases was 76 cases/year during the study period. Overall, the HCIR was 7.95 cases/100 000 PY. It was higher among females (9.1 cases/100 000 PY). According to age groups, the highest HCIR was observed among patients aged 60 years and above (16.5 cases/100 000 PY). The hospital crude mortality rate was 0.13/100 000 inhabitants/year. It was 0.14/100 000 inhabitants/year for females and 0.06/100 000 inhabitants/ year for males (Table 1). A significant and low increase of RHD patients’ mean age was notified from 2003 to 2016 (Rho = 0.2; p < 0.001).

Chronological Trends and Projection of RHD New Cases up to 2026
Joinpoint regression analysis showed a decreasing trend in overall RHD new cases from 2003 to 2016, but no significant change was noted (APC = -4.4%; 95%CI [-10.4; 2.1%]; p = 0.2). The same results were applied for males (APC = -5.4%; 95%CI [-12.4; 2.2%]; p = 0.1) and females (APC = -3.8%; 95%CI [-9.3; 2.2%]; p = 0.2) (Figure 1). As for age-specific curves, trends analysis of RHD new cases showed a significant decline from 2003 to 2016 for patients aged under 25 years (APC = -15.95%; 95%CI [-21.8; -9.6%]; p < 0.001). For patients aged between 25 and 39 years (APC = -6.8%; 95%CI [-13.7; 0.7%]; p = 0.1) and those aged between 40 and 59 years (APC = -2.69%; 95%CI [-9.5; 4.7%]; p = 0.4), we noted a declining trend, but without significant change over time, while for those aged 60 years and above, trends curve remain stable from 2003 to 2016, with an APC of 0.64% (95%CI [-6.3; 8.1%]; p = 0.9) (Figure 2).

The number of new RHD cases predicted by the age-period-cohort model would attend 35 cases by 2026 ([LCrI = 27, UCrI = 46]). According to age groups, the mean estimated number of new RHD cases would attend 5 cases up to 2026 ([LCrI = 3, UCrI = 6) for patients aged under 25 years, 8 cases([LCrI = 6, UCrI = 11]) for those aged between 25 and 39 years, 14 new cases ([LCrI = 11, UCrI = 19]) for those aged between 40 and 59 years and 8 new cases ([LCrI = 6, UCrI = 11]) for those aged 60 years and above (Figure 3).

Discussion
Rheumatic heart disease is one of the most common and preventable acquired heart diseases. In highincome countries, the burden of RHD decreased markedly during the 20th century, attributable to improved living conditions associated with socioeconomic development and advances in medical management [6]. In the developing countries, militating against RHD remains a daily challenge, due to a poor awareness of physicians as well as patients of the importance of penicillin usage for acute rheumatic fever. The estimated number of RHD in 2015 was 33194900 cases that occurred in countries with an endemic pattern of disease, while only 221600 cases were in countries with a non-endemic pattern [5].The highest prevalence rates were in Western Pacific Region Oceania, followed by central sub-Saharan Africa and South Asia [5,6]. In 2015, the countries with the largest estimated numbers of cases of RHD were India (13.17 million cases), China (7.07 million), Pakistan (2.25 million), Indonesia (1.18 million) and the Democratic Republic of the Congo (805000), together accounting for 73% of global cases [5]. At a local level, a previous survey conducted on a hospital-based population in Central-east of Tunisia during the same period reported a total number of 1063 hospitalizations for RHD, with an incidence of 10.97/100 000 PY [7], which was similar to our findings.

According to WHO, the RHD-related mortality varies widely, from1.8/105inhabitants in the region of the Americas to 7.6/105 inhabitants in the region of South-East Asia, which was lower than our rates [8], whereas other studies showed similar case-fatality rates to our study, ranging from 1 to 1.5% [1,7].

Another important finding in our study suggested the predominance of RHD among women. Various other studies also revealed higher rates among females [9,10]. It has been reported that female gender was associated with a higher risk of having borderline RHD, with an OR of 1.9 [11]. This might be the result of innate susceptibility or an increased exposure to group A streptococcus probably due to a greater involvement of women in child rearing.

During the study period, the diagnosis age of patients with RHD has significantly increased over time and the highest HCIR was observed among elderly (age ≥ 60 year). These results were in line with other Korean studies reporting higher prevalence rates of RHD among older age groups, resulting in high disease burden and increasing health-care costs [12,13]. These findings supported the hypothesis suggesting that RHD can occur at any age, even at an advanced age, which reflected the under-screening rate of RHD during childhood. Therefore, future designs of cardiovascular healthcare services in countries with a rapidly aging population should be urgently implemented. The lowest rate was observed among younger patients (age < 25 years). These results were not contrasting with previous researches reporting high prevalence of RHD among children and adolescents [14-16], but highlighted the long latency period between the pharyngeal infection with Group A β hemolytic Streptococcus, the acute rheumatic fever, followed by asymptomatic involvement of the heart valves fibrosis and the beginning of clinical signs in our population. Thus, we should emphasize on the value of school screening programs for detecting asymptomatic RHD as an effective alternative to morbidity registers that are based solely on clinical encounters.

The mitral valve was the most common localization of the RHD. The same findings have also been shown in previous studies [17-19]. An Indian survey reported that mitral valve disease accounted for 86% of cases, aortic valve involvement was seen in 7% of cases [18]. It has been reported that the tricuspid valve is rarely affected by rheumatic carditis, but functional tricuspid regurgitation may accompany mitral valve disease [17].

The number of cases in developed countries have reduced drastically [5]. However, the same did not apply to developing countries, like those in sub-Saharan Africa. In Nigeria, the prevalence of RHD has increased from 2.9 to 9.8% and from 17.4 to 57.7% for the structural and acquired heart diseases, respectively [19]. On the other hand, a significant decline was noted in a previous Tunisian study conducted from 2000 to 2013 [7], which was consistent with our study. This might be explicated by the implementation of acute articular rheumatism monitoring program in Tunisia since 1980, the delivery of penicillin for the treatment of sore throat, notably at an early age, and the health education of the parents concerning the therapeutic observance during the treatment of angina, which provided a strong explanation for the significant decrease of RHD among patients under 25 years. Furthermore, our country is undergoing an epidemiological transition during the last decades, characterized by the predominance of chronic diseases, mainly ischemic heart disease at the expense of infectious diseases.

Conclusion
Our study provided original findings about the substantial burden of RHD in Southern Tunisia. In spite of declining curves in the hospitalizations of RHD, it still remains a major public health problem for both sexes and in all age-groups. The decreasing trend was mainly observed among young patients, with a significant change over time. A reduction in social inequalities and unified global approach based on early screening of silent RHD at an early age represent the cornerstones of community-based prevention of RHD, particularly in low-resources countries.

Supplementary Files (if applicable)
Not applicable

Acknowledgements
No financial grant.

Conflicts of Interest
No conflicts of interest.

Bibliography

1. Carapetis, J. R., Steer, A. C., Mulholland, E. K. & Weber, M. (2005). The global burden of group A streptococcal diseases. Lancet Infect Dis., 5(11), 685-694.
2. Ralph, A. P. & Carapetis, J. R. (2012). Group A Streptococcal Diseases and Their Global Burden. Current Topics in Microbiology and Immunology, 368, 1-27.
3. WHO. Rheumatic fever and rheumatic heart disease. World Health Organization, 2013.
4. Murray, C. J. L., Vos, T., Lozano, R., Naghavi, M., Flaxman, A. D., Michaud, C., et al. (2012). Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet, 380(9859), 2197-2223.
5. Watkins, D. A., Johnson, C. O., Colquhoun, S. M., Karthikeyan, G., Beaton, A., et al. (2017). Global, Regional, and National Burden of Rheumatic Heart Disease, 1990-2015. N Engl J Med., 377(8), 713-722.
6. Abouzeid, M., Katzenellenbogen, J., Wyber, R., Watkins, D., Johnson, T. D. & Carapetis, J. (2017). Rheumatic heart disease across the Western Pacific: not just a Pacific Island problem. Heart Asia. British Medical Journal Publishing Group, 9(2), e010948.
7. Sriha Belguith, A., Koubaa Abdelkafi, A., El Mhamdi, S., ben fredj, M., Abroug, H., Ben Salah, A., et al. (2017). Rheumatic heart disease in a developing country: Incidence and trend (Monastir; Tunisia: 2000-2013). Int J Cardiol., 228, 628-632.
8. Sorour, K. A. (2014). Rheumatic heart disease in Egypt: Gloomy past and promising future. Egypt Hear J. Elsevier., 66(2), 139-142.
9. Shrestha, N. R., Karki, P., Mahto, R., Gurung, K., Pandey, N., et al. (2016). Prevalence of Subclinical Rheumatic Heart Disease in Eastern Nepal. JAMA Cardiol., American Medical Association, 1(1), 89-96.
10. Padmavati, S. (2001). Rheumatic heart disease: prevalence and preventive measures in the Indian subcontinent. Keywords: rheumatic heart disease; rheumatic fever. Heart, 86(2), 127.
11. Corsenac, P., Heenan, R. C., Roth, A., Rouchon, B., Guillot, N. & Hoy, D. (2016). An epidemiological study to assess the true incidence and prevalence of rheumatic heart disease and acute rheumatic fever in New Caledonian school children. J Paediatr Child Health, 52(7), 739-744.
12. Seo, H. Y., Yoon, S. J., Kim, E. J., Oh, I. H., Lee, Y. H. & Kim, Y. A. (2013). The economic burden of rheumatic heart disease in South Korea. Rheumatol Int., 33(6), 1505-1510.
13. Jang, S. Y., Ju, E. Y., Seo, S. R., Choi, J. Y., Park, S. J., et al. (2014). Changes in the etiology of valvular heart disease in the rapidly aging Korean population. Int J Cardiol., 174(2), 355-359.
14. Bhaya, M., Panwar, S., Beniwal, R. & Panwar, R. B. (2010). High Prevalence of Rheumatic Heart Disease Detected by Echocardiography in School Children. Echocardiography, 27(4), 448-453.
15. Sadiq, M., Islam, K., Abid, R., Latif, F., Rehman, A. U., et al. (2008). Prevalence of rheumatic heart disease in school children of urban Lahore. Heart, 95(5), 353-357.
16. Katzenellenbogen, J. M., Ralph, A. P., Wyber, R. & Carapetis, J. R. (2017). Rheumatic heart disease: infectious disease origin, chronic care approach. BMC Health Serv Res., 17(1), 793.
17. Sani, M. U., Karaye, K. M. & Borodo, M. M. (2007). Prevalence and pattern of rheumatic heart disease in the Nigerian savannah: an echocardiographic study. Cardiovasc J Afr., 18(5), 295-299.
18. Sriharibabu, M., Himabindu, Y. & Kabir, Z. (2013). Rheumatic heart disease in rural south India: A clinico-observational study. J Cardiovasc Dis Res., 4(1), 25-29.
19. Animasahun, B. A., Madise Wobo, A. D., Itiola, A. Y., Adekunle, M. O., Kusimo, O. Y. & Thomas, F. B. (2018). The burden of rheumatic heart disease among children in Lagos: how are we fairing? Pan Afr Med J., 29(150).