It was estimated recently that world wide 15.6 million people have rheumatic heart disease and that there are 470 000 new cases of rheumatic fever and 233 000 deaths attributable to rheumatic fever or rheumatic heart disease each year.1 These are conservative estimates, especially if echocardiographic screening is used when the actual figures are likely to be substantially higher.2 Almost all these cases and deaths occur in developing countries like Pakistan.

In Pakistan, like many developing countries, rapid industrialisation has brought a population shift from rural to urban areas, leading to crowded urban slums.3 Rheumatic fever and rheumatic heart disease (RHD) still have a very high prevalence and RHD is one of the leading causes of premature death and disability in Pakistan.4 The initial community-based studies of RHD prevalence conducted in the 1960s and 1970s5 6 showed a prevalence of RHD of about 6 in 1000. These studies, however, were non-randomised and not based on echocardiographic diagnosis. A more recent community-based survey from rural Pakistan which was randomised and based on echocardiography has shown a similar prevalence of 5.7 in 1000.7 The other epidemiological data are limited to hospital-based reports.8 9 As tertiary care centres are located in urban areas, only a few people mange to seek such treatment and such hospital-based data are unlikely to reflect the true prevalence in the vulnerable population of the community.

The criticism of school-based studies in countries like Pakistan has been low enrolment rates and frequent absenteeism. Despite these limitations, school-based studies are the most reliable studies possible in these developing countries and reflect the epidemiology of RHD in the most vulnerable population. The previous school population studies were either performed before the echocardiography era (1960s and 1970s) and lacked diagnostic precession or were not representative of the majority population (table 1).10^–13 There has been a gradual shift of population in Pakistan from rural to urban areas,14 and these urban slums now probably bear the greatest burden of RHD mortality and disability. We conducted this survey to determine the school survey-based prevalence of RHD in the urban slums of the city of Lahore, a representative second largest city in Pakistan, capital of the most populous province of Punjab.

METHODS

Setting

Lahore is the second largest cosmopolitan city of Pakistan and is the capital of the province of Punjab—the most populous of the four provinces of the country. The climate is generally hot as the main season is summer (April to September). Winter is generally mild. The monsoons are at their peak during July and August, and during these 2 months there is more than half of the annual rainfall. According to the 1998 census,14 Lahore’s population was nearly 6.8 million. Mid-2006 government estimates now put the population at somewhere around 10 million, of whom 2.5 million (25%) reside in urban slums and so called Kachi Abadis (houses constructed with semipermanent materials). A typical family’s income is Pakistani 3000 rupees (US$50) a month. The average family size is 7.12 people. The overall literacy rate is 54%, of which 66% men and 42% women are literate. Primary education is free and these Lahore Municipal Corporation schools are located conveniently within these areas.

Sampling technique and survey methods

The study was carried out after prospective planning by a group of cardiologists, which included establishing protocols for prospectively defined clinical and then echocardiographic criteria for the diagnosis of RHD. The study was carried out from May 2001 to April 2002 after approval by the hospital ethical committee. The investigators approached the principals of all 70 schools of the Lahore Municipal Corporation and they agreed to participate.

A total of 24 980 pupils from 70 schools were randomly selected from among the children 5 through 15 years of age. The overall sample was equally distributed among nine class levels, with 2770–2780 children in each class level, in order to ensure an even age distribution. For each class level, classrooms were selected by a systematic randomisation procedure with Epi Info software. In each selected classroom, all selected children underwent comprehensive clinical physical examination within the school for signs of a cardiac disease be it RHD or congenital heart disease (CHD). Absentees were noted and second visits were made in an attempt to examine all of them.

Careful cardiac auscultation was performed by a cardiologist (KI, RA) with the patient in the supine and left lateral decubitus positions. All children in whom a cardiac lesion was suspected by clinical examination (presence of heart murmur or abnormal heart sounds, peripheral stigmata-like pulse abnormalities, cyanosis or clubbing) or who had a history of rheumatic fever even in the absence of audible murmur were brought to the tertiary care centre, Punjab Institute of Cardiology. These children then underwent scanning with a Toshiba Power vision and Toshiba SSH-60A. The echocardiography was performed by cardiologists experienced in paediatric echocardiography (MS, AUR and FL). The echocardiographic criteria were agreed by all cardiologists before the scans were performed.

Only left-sided valves were examined for features of RHD; mild tricuspid regurgitation and pulmonary regurgitation, although often seen, were disregarded. RHD was defined by the presence of any definite evidence of mitral or aortic valve regurgitation seen in two planes by Doppler echocardiography, accompanied by classical morphological abnormalities of the regurgitant valve like restricted leaflet mobility, focal or generalised valvular thickening and abnormal subvalvular thickening.15 16 Multiple cross-sectional views were taken from apical, parasternal and subcostal positions according to recommendations of the American Society of Echocardiography.15 16 Criteria for pathological valvular regurgitation agreed upon included:

• colour jet seen in at least two planes;

• mosaic colour jet (mosaic posterolateral jet in at least two planes was considered pathologic for the mitral valve);

• length of colour jet ⩾1 cm.

Continuous or pulse-wave Doppler examination of the regurgitant flow needed to be holosystolic for mitral regurgitation, or holodiastolic for aortic regurgitation, with peak velocities exceeding 2.5 m/s.

It was compulsory for one of the parents or a guardian to accompany the child when brought to the tertiary referral centre for detailed echocardiography. At this stage we gathered the socioeconomic and demographic details for each child confirmed to have RHD. The sociodemographic variables gathered were the age, sex, occupation and literacy (based on ability to read and write a simple letter). The socioeconomic status was based on occupation, salary, ownership of common household items and the type of house. We measured crowding based on the number of household members and number of rooms in the house, ranging from one (single room with five or more people) to three (three rooms with fewer than five people or more than three rooms).

The detailed information for all patients with echocardiographically proven RHD included a structured medical history, physical examination and laboratory tests when indicated (12-lead ECG, chest x-ray examination, complete blood count, acute phase reactants like erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) and antistreptolysin O (ASOT) to seek evidence of previous streptococcal infection). We collected details of any manifestations of rheumatic fever from each patient and their accompanying family member according to the updated 1992 Jones criteria of the American Heart Association task force on rheumatic fever/rheumatic heart disease17 as well as a history of joint swelling, arthralgia, fever, rheumatic prophylaxis and cardiac surgery. Each child in whom rheumatic or congenital heart disease was diagnosed by echocardiography was followed up at the tertiary referral centre with clinical and echocardiographic review, initiation of secondary rheumatic prophylaxis and surgical or interventional treatment where clinically indicated.

Statistical analysis

Data were entered in the computer using ISSPS version 13, and the same program was used to perform statistical analysis. For quantitative variables, means and median were calculated, while for qualitative variables, proportions along with 95% confidence intervals were calculated (Epi Info 6 computer software was used to calculate 95% confidence intervals).

RESULTS

We surveyed a total of 70 schools and a total of 24 980 children between the ages of 5 and 15 years. Of these 24 980 children screened, 1529 (6%) were suspected as having a cardiac lesion clinically. Of these, 1489 (97.4%) could be brought to the tertiary care hospital for detailed echocardiography. Of these, 986 (66.2%) had a cardiac lesion on echocardiography while others were normal and excluded. A history of symptoms of rheumatic fever was elicited from 196/546 (36%) patients with confirmed RHD. Among those with confirmed heart disease, 546 (55%) had RHD while 440 (45%) had congenital heart disease or other heart diseases (fig 1). The prevalence of RHD was therefore, 21.9/1000 school children screened (95% CI 20.1 to 23.7).

The mean (SD) age of the children with RHD was 10.7 (2.6) years (95% CI 10.5 to 11.0) and 61.5% were female. The median age was 11 years with interquartile range of 4. Table 2 shows the sociodemographic data. The population we surveyed lived in crowded conditions, each household having an average of seven members living in 2.3 rooms. The families were poor, 81% earning less than 3000 rupees (US$50) a month (median income of 2600 rupees (US$45) a month). The nutritional status was also assessed. Height and weight were plotted on the centile charts for Pakistani children and 64% of all children were below the third centile for height and weight.

The mitral valve was affected in most cases, with mixed lesion of mitral regurgitation and mitral stenosis being the commonest lesion (30%), followed by isolated mitral stenosis (28%) and isolated mitral regurgitation (22%). A mixed lesion of mitral regurgitation and aortic regurgitation (8%) was less common and so was the occurrence of isolated aortic regurgitation (3.1%) (table 3). Mitral stenosis was seen at a much younger age (mean (SD) 10.8 (2.4) years) and usually severe. These children were grouped into two groups by their age and prevalence of various lesions and table 4 describes their relative rates by age. Most children (82%) were asymptomatic, and in 92.5% RHD was diagnosed for the first time during the school survey. Less than 2% of the affected children were taking secondary rheumatic prophylaxis.

Of children with other heart diseases, 417 had CHD. The prevalence of CHD hence was 16.7/1000 (95% CI 15.1 to 18.4). Most of these had a mild form of the disease. Isolated ventricular septal defect (VSD) was the commonest lesion (59%), followed by pulmonary stenosis (12%), atrial septal defect (8%), valvular aortic stenosis (7%), patent ductus arteriosus (5%), tetralogy of Fallot (3%), dilated cardiomyopathy (2%), hypertrophic cardiomyopathy (1%) and others including mitral valve prolapse (3%). Almost one-quarter of these children needed an interventional or surgical treatment.

DISCUSSION

There is a very high prevalence of RHD among the school children of urban slums and adjoining semiurban Kachi Abadis of Lahore. This is one of the highest reported in the echocardiography era18 except where echocardiography itself was used as a screening tool.2

Three groups of factors seem to affect the distribution of RHD globally: (a) environmental factors; (b) genetics and the host; (c) the organism. The environmental factors include socioeconomic status, urbanisation, overcrowding, nutrition and access to medical care. The data in our study were collected from a poor socioeconomic section of the society with overall prevailing poverty in the area. There was overcrowding with each household having an average of seven members living in 2.3 rooms. Despite some conflicting data, the link between socioeconomic indicators and RHD is well established. In Kinshasa, the prevalence of RHD was found to be far higher in semiurban areas (22.2 per 1000) than that in urban areas (4 per 1000).19 In these semiurban areas, people lived in crowded conditions, with 81.1% of subjects in houses with more than eight people. In Rajasthan, India, the prevalence of RHD in the lowest socioeconomic status group was 3.9/1000, in the middle socioeconomic status group 2.1/1000 and in the highest socioeconomic status group there was no RHD.20

The disease occurs in more than one family member and clinicians have long appreciated the higher familial incidence but have not been able to demonstrate a specific genetic pattern. In our study the other family members were not screened and RHD was found in three siblings studying in the same school. More substantial evidence for a genetic association was provided by Khanna and associates, who reported that B-cell alloantigens, designated D8–17, were present in 99% of patients with rheumatic fever.21 Further support for the role of genetic factors in susceptibility was provided by the associations of this disease with inheritance of the HLA major histocompatibility antigens. The genotype DRB1*06-DQB1*0602-8 may be associated with protection against rheumatic fever and the development of RHD. The severity of rheumatic fever is probably associated with the DRB1*07 allele and development of certain RHD may be dependent on specific DQ alleles.22

In the absence of a clear genetic explanation, the differences in rates between populations may be due to factors related to the organism. Although the emergence of virulent group A streptococci is thought to explain recent outbreaks of rheumatic fever in the USA,23 there is little evidence of the circulation of particularly virulent strains in many developing countries. In Aboriginal groups in the Northern Territory in Australia, who are known to have high rates of rheumatic fever, serotypes of group A streptococci traditionally associated with rheumatic fever are not commonly found.24 There are no concrete data on this issue from the Indian subcontinent.

Cross-sectional school surveys reported world wide have relied on careful clinical examination of school-age children, with confirmation of clinically suspected cases by echocardiography. Such surveys show current prevalence rates of RHD of approximately 1–15 cases per 1000 among school-age children in developing countries.18 25 Only two surveys have included echocardiography of all surveyed children.2 26 In one study,26 conducted in Kenya, the prevalence of echocardiographically detected trivial mitral regurgitation was 62 per 1000 and the prevalence of clinically and echocardiographically detected RHD was only 2.7/1000 school children. These investigators did not present any data on valve morphology or the characteristics of the regurgitant jet; hence it was not clear whether the regurgitation was physiological or was due to rheumatic or congenital disease. In the recent more structured study2 clinical examination detected RHD that was confirmed by echocardiography in eight of 3677 children in Cambodia (2.2/1000) and five of 2170 children in Mozambique (2.3/1000). In contrast, echocardiographic screening detected 79 cases of RHD in Cambodia and 66 cases in Mozambique, corresponding to prevalence rates of 21.5/1000 and 30.4/1000, respectively. This study documented a much higher prevalence (10 times higher than clinical screening) of RHD in school-age children. One limitation of our study is that no echocardiography was carried out even in a sample of children without a cardiac murmur. However, mass clinical screening is more cost effective than echocardiographic screening.

Age of onset and severity of rheumatic fever in the Indian subcontinent has been of special interest. Roy et al first introduced the term “juvenile mitral stenosis” and reported a high (26.7%) prevalence of severe RHD in patients aged <20 years.27 Most patients in our study were young (mean age 10.7 years) and the incidence of mitral stenosis was also disproportionately high (28%). The underlying reasons may relate to recurrent, unrecognised or untreated streptococcal infections, lack of secondary prophylaxis, unusual strains of streptococci, malnutrition or to genetic and ethnic differences in patients from this part of the world.28 29 The specific role of each factor has not been defined.

The prevalence of RHD in urban slums of Lahore is an unusually high rate from a disadvantaged population and a more realistic average for the country may be somewhere in between the reported low and high rates. Another factor also confounds these high prevalence data. There are selected small urban pockets within these cities which have a considerably higher per capita income and human development index. These “islands” of affluence amidst ocean of “deprivation” are likely to have a very low prevalence of RHD. School surveys done in Pakistan under the WHO Global Programme for the Prevention of rheumatic fever/RHD did not take that into account and prevalence ranged from 0.6 to 7 per 1000 depending upon the area selected.13 One of the most disturbing findings in our study was that very few (<8%) of the children were aware of the diagnosis and only 2% were having benzathine penicillin injections.

There is now renewed interest in rheumatic fever and RHD given their persistence and enormous socioeconomic burden, and a review article by Nkomo has given an account of this problem in Africa.30 In response to the World Health Organization proposed guidelines (2004) for the control of rheumatic fever and RHD, the Pan African Society of Cardiology (PASCAR) convened the first All Africa Workshop on rheumatic fever and RHD on 15–16 October 2005, and adopted an action plan called the Awareness Surveillance Advocacy Prevention (ASAP) Programme.31 This is meant to be simple, yet comprehensive, to be adopted in part or whole by the continent’s national departments of health or non-governmental organisations. The ASAP programme, while recognising the barriers to implementation, aims to apply best known practices of what works by (a) raising awareness about the disease among child caregivers, teachers and healthcare workers; (b) establishing surveillance systems modelled after the “WHO STEP wise approach” used to collect epidemiological data on risk factors for non-communicable diseases in developing countries; (c) championing advocacy for rheumatic fever and RHD, which have been neglected because of their virtual disappearance in wealthier countries; (d) implementing primary and secondary prevention strategies that have been proved to be effective.

These data paint an alarming picture of the high prevalence of RHD in a resource-constrained environment where access to healthcare is limited. The available and potential control measures for rheumatic fever and RHD are “primordial prevention” (housing and hygiene), “primary prevention” (sore throat treatment, vaccine-unavailable), “secondary prevention” (secondary prophylaxis) and “tertiary prevention” (medical treatment, surgery or intervention, anticoagulation). Of these, only one, secondary prophylaxis has been proved to be cost effective and practical. Yet, Pakistan like most developing countries still does not have effective control programmes, in particular, secondary-prophylaxis programmes, and rheumatic fever and RHD remain major unresolved health problems.