Objectives This scoping review sought to summarise available data on the prevalence, aetiology, diagnosis, treatment and outcome of pericardial disease in Africa.
Methods We searched PubMed, Scopus and African Journals Online from 1 January 1967 to 30 July 2017 to identify all studies published on the prevalence, aetiologies, diagnosis, treatment and outcomes of pericardial diseases in adults residing in Africa.
Results 36 studies were included. The prevalence of pericardial diseases varies widely according to the population of interest: about 1.1% among people with cardiac complaints, between 3.3% and 6.8% among two large cohorts of patients with heart failure and up to 46.5% in an HIV-infected population with cardiac symptoms. Tuberculosis is the most frequent cause of pericardial diseases in both HIV-uninfected and HIV-infected populations. Patients with tuberculous pericarditis present mostly with effusive pericarditis (79.5%), effusive constrictive pericarditis (15.1%) and myopericarditis (13%); a large proportion of them (up to 20%) present in cardiac tamponade. The aetiological diagnosis of pericardial diseases is challenging in African resource-limited settings, especially for tuberculous pericarditis for which the diagnosis is not definite in many cases. The outcome of these diseases remains poor, with mortality rates between 18% and 25% despite seemingly appropriate treatment approaches. Mortality is highest among patients with tuberculous pericarditis especially those coinfected with HIV.
Conclusion Pericardial diseases are a significant cause of morbidity and mortality in Africa, especially in HIV-infected individuals. Tuberculosis is the most frequent cause of pericardial diseases, and it is associated with poor outcomes.
- pericardial disease
- cardiac tamponade
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Pericardial diseases represent a wide range of clinicopathological entities characterised by damage to the pericardium. Their clinical manifestations vary significantly and range from asymptomatic to life-threatening presentations. The most common forms of pericardial diseases include acute pericarditis, effusive pericarditis with or without tamponade, pericardial constriction and effusive constrictive pericarditis. Pericardial diseases are a significant cause of morbidity and mortality, accounting for about 4%–13% of causes of heart failure in some studies in sub-Saharan Africa.1 2 Although the prevalence of pericardial disease-related heart failure is reported to be relatively low on the African continent, it is associated with very high mortality rates. For instance, the 6-month mortality rate for tuberculous pericarditis was 40% among persons living with HIV/AIDS in a recent Pan African registry.3
The HIV epidemic in sub-Saharan Africa has led to a surge in the incidence of pericardial diseases over the past few decades, driven predominantly by an increase in tuberculous pericarditis.3 4 This has resulted in greater attention and interest in the problem by clinicians and researchers alike. Alongside the increasing number of epidemiological studies on pericardial diseases,1 some clinical trials have been conducted or are ongoing to support evidence-based management of these diseases.5 However, there have been few attempts to synthesise and summarise the available published clinical research experience. This scoping systematic review summarises available data on the prevalence, aetiology, treatment and outcome of pericardial disease in Africa.
This scoping review was conducted according to the five-step approach proposed by Arsksey and O’Malley.6
We searched PubMed/MEDLINE, Scopus and African Journals Online from 1 January 1967 to 30 July 2017, without language restriction, to identify all published studies providing data on the prevalence, aetiologies, diagnosis, treatment and outcomes of pericardial diseases in African-based populations. The search strategies are available in the appendix (online supplementary table 1). We manually searched the reference list of all relevant articles and reviews to identify additional articles.
Supplementary file 1
Selection of studies for inclusion in the review
We included all cross-sectional, cohort, meta-analysis and case–control studies reporting relevant data on pericardial diseases in African populations. We excluded editorials, commentaries, letters to the editor, case reports, studies that included non-African populations or African populations residing outside Africa. Data from relevant systematic reviews were also considered. For duplicate publications, we considered only the most recent and/or comprehensive study with the largest sample size.
Two investigators (JJN and VNA) independently screened the titles and abstracts of articles retrieved from literature search, and the full-texts of potentially eligible articles were obtained and further assessed for final inclusion (Supplementary figure 1). Disagreements were resolved through consensus.
Data on the prevalence, aetiologies, diagnosis, treatment and outcome of pericardial diseases were extracted by four investigators (ALN, UFN, JRN and AK) with the aid of prestructured data abstraction sheets. All the extracted data were cross-checked for consistency and correctness by a fifth investigator (JJN). Data were extracted on the last name of the first author, publication year, study country, study design, data collection (prospective and retrospective), study setting (community based vs hospital based), study area (rural, urban or semiurban), study population, male proportion, mean age, age range, proportion of patients with HIV, sample size, prevalence or incidence of pericardial diseases, distribution of aetiologies, clinical patterns, treatment and outcomes. Due to high heterogeneity across studies, we opted for a narrative summary of our findings.
We identified 1512 records from bibliographic searches. After screening titles and abstracts, we retained and assessed 151 full-text papers for eligibility. Finally, 36 studies were included7–42 (appendix, online supplementary figure 1).
Prevalence of pericardial diseases
Seventeen studies reported data on the prevalence of pericardial diseases in various populations7–23 (table 1). For clarity of purpose, we sought to determine the prevalence of pericardial disease (PD) in studies of three major populations: (1) participants with confirmed or suspected heart failure; (2) persons living with HIV/AIDS; and (3) patients with multisystem conditions such as systemic lupus erythematosus (SLE).
In patients with heart failure, pericardial disease was found to be the cause in 3.3%–13.2% of cases.7–11 Specifically, in The Sub-Saharan Africa Survey of Heart Failure study that involved 1006 cases of heart failure from several sub-Saharan African countries, 6.8% of patients had a pericardial effusion with tamponade causing acute heart failure.7 Another study from Nigeria including 452 patients with acute heart failure found a prevalence of pericardial disease of 3.3%.8 Among 2501 patients presenting with cardiac complaints in Nigeria, 1.1% had a pericardial disease.13 A similar prevalence rate was found in Burkina Faso (1.1%, n=945) and Togo (2%, n=1896) among patients admitted to cardiology units and among patients in Nigeria who underwent echocardiography to investigate symptoms of dyspnoea and heart failure (1.8%, n=1441).13–15
The prevalence of pericardial diseases has also been investigated in a heterogeneous group of studies investigating causes of heart disease in people living with HIV/AIDS with and without cardiac symptoms from around the continent.16–20 In Mali, among 821 HIV-infected patients, 1.9% were found to have a pericarditis.16 High frequencies of pericardial effusion or pericarditis were found among HIV-infected patients presenting with cardiac symptoms or admitted to a cardiology unit as reported in three studies in Cameroon (46.5%),17 in Tanzania (41.1%)18 and Burkina Faso (35.4%),19 although these estimates were derived from small size populations.
Pericardial disease may occur as a complication of a number of multisystem diseases, but the frequency of pericardial involvement in these conditions in the African setting is not known. These include malignancies, SLE, hypothyroidism, chronic renal failure and tuberculosis. Three studies reported on the prevalence of pericarditis among patients with SLE. This ranged from 18% (n=100) based on a clinical diagnosis to 77.8% (n=63) of SLE patients with pericardial thickening.21–23 Only one study reported on the prevalence of pericardial involvement among patients with tuberculosis.20 In this study conducted in a tuberculosis-specialised hospital in Cameroon, the authors observed a prevalence of pericarditis of 0.9% among 984 cases of tuberculosis, irrespective of the HIV status.
Aetiology of pericardial diseases
Nine studies reported on the aetiologies of pericardial diseases in Africa. Most of these had a small sample size of less 100 participants and few described the investigations used or methods of establishing the diagnosis13 16 24–31 (table 2). The oldest report on the aetiology hailed from Zimbabwe and presented data collected between 1967 and 1971. It identified tuberculosis as the predominant cause of effusive pericarditis (64.9%).24 The largest study was a South African single-centre hospital experience of 233 patients with large pericardial effusions. About 36.1% of the participants were infected with HIV, and tuberculosis was the most common cause (69.5%), followed by malignancies (9.4%).25 In a Tanzanian study, tuberculosis was the cause of all 14 (100%) pericardial effusions among persons living with HIV and AIDS.27 Overall, studies have demonstrated that a cause could be found in between 9% and 65% of cases with pericardial disease. Tuberculosis, followed by malignancies and rheumatological diseases were the most common causes of effusive pericarditis in Africa.13 16 28–30 The aetiology remained unknown in a large proportion of cases, including 48.6% of participants in a study from Mali.31 Finally, only one study conducted in South Africa reported specifically on the aetiology in patients with a diagnosis of constrictive pericarditis. It reported that 90.9% of the 121 cases (11.6% with HIV infection) undergoing pericardiectomy for constrictive pericarditis were due to tuberculosis.26
Clinical patterns of pericardial diseases
The forms of pericardial diseases were reported in six studies15 16 30–33 (table 3). In the multinational Investigation of the Management of Pericarditis (IMPI) registry, a multicentre, prospective, observational study of patients with presumed tuberculous pericarditis who were admitted consecutively to 15 referral hospitals in Cameroon, Nigeria and South Africa,34 35 patients with tuberculous pericarditis presented mostly with effusive pericarditis (79.5%) and in 15.1% and 13% of cases with effusive constrictive pericarditis and myopericarditis, respectively. Furthermore, 29.2% of patients presented with haemodynamic instability and 20.4% with life-threatening cardiac tamponade.32 Specifically, patients with clinical HIV disease were more likely to present with dyspnoea and electrocardiographic features of myopericarditis.32 At 6-month follow-up, those with HIV-infection had a lower incidence of constriction.33 In another South African study of patients with TB pericarditis that looked more closely for evidence of myocardial injury, the proportion of patients with tuberculous myopericarditis was much higher at 53.1%.34 In a study of pericarditis in HIV-infected patients from Mali, the majority of participants presented with effusive pericarditis (55.1%) and 20.4% had evidence of cardiac tamponade.16
Recent advances on the diagnosis and management: emphasis on tuberculous pericarditis
As shown in table 2, an aetiology was not found in about 15%–20% of cases of pericardial diseases. A definitive diagnosis of tuberculous pericarditis is very challenging owing to a suboptimal accuracy of current diagnostic tests. Consequently, some studies have been conducted to explore the accuracy of new diagnostic tests for tuberculous pericarditis, including the quantitative PCR test (Xpert MTB/RIF), the unstimulated interferon gamma (uIFNγ), pericardial adenosine deaminase (ADA) and pericardial and urinary lipoarabinomannan (LAM). Studies conducted in patients enrolled in the IMPI registry and trial have significantly contributed to the body of knowledge on the topic. One study that evaluated the diagnostic accuracy of the Xpert MTB/RIF test compared with pericardial ADA and uIFNγ in suspected tuberculous pericarditis showed that uIFNγ assay may be the optimal first line test for the diagnosis of TB pericarditis. Furthermore, pericardial fluid Xpert MTB/RIF, when combined with either ADA or uIFNγ, offers higher sensitivity and specificity for the diagnosis of tuberculous pericarditis.43 Another study showed a low sensitivity but high specificity of urinary or pericardial LAM for the diagnosis of tuberculous pericarditis, with an increased sensitivity of urinary LAM in HIV-infected patients with CD4 ≤100 cells/mm3.44
The management of tuberculosis pericarditis usually entails an interplay between antituberculous therapy, pericardial drainage and pericardiectomy (table 4).12 15 23–25 34–41 Some adjunctive therapies such as corticosteroids have been used in order to improve survival and pericarditis-related outcomes. It was postulated that the use of adjunctive glucocorticoid in the treatment of tuberculous pericarditis might reduce the inflammatory response to tuberculoproteins, leading to a decrease in the occurrence of pericardial effusion, tamponade and constrictive pericarditis and therefore reduce mortality.45 The most important trial on the use of corticosteroid in tuberculous pericarditis, the IMPI trial, failed to show a significant effect on the composite of death, cardiac tamponade requiring pericardiocentesis or constrictive pericarditis.46 A recent Cochrane review on the interventions for treating tuberculous pericarditis suggested that in HIV-uninfected individuals, glucocorticoid probably reduce deaths from pericarditis (risk ratio (RR) 0.39, 95% CI 0.19 to 0.80), even though there was no significant effect on all-cause mortality (RR 0.80, 95% CI 0.59 to 1.09) and the need for repeat pericardiocentesis (RR 0.85, 95% CI 0.70 to 1.04).47 In people infected with HIV, corticosteroids did not significantly reduce the occurrence of pericardial constriction (RR 0.55, 95% CI 0.26 to 1.16), the need for repeat pericardiocentesis (RR 1.02, 95% CI 0.89 to 1.18) and all-cause mortality.47 It is worth mentioning that all the studies included in this review were conducted in Africa. In the IMPI trial, specifically, corticosteroids were associated with a significant increase in the occurrence cancer (HR 3.27; 95% CI 1.07 to 10.03), especially HIV-associated cancers.46 The IMPI trial also showed that irrespective of HIV status, immunotherapy with Mycobacterium indicus pranii has no clinical benefit but rather increases the incidence of cancer.46
Outcomes of pericardial diseases
Fourteen studies investigated the outcomes of pericardial disease in sub-Saharan Africa13 16 24–26 34–42 (table 4). Most of these studies involved patients with tuberculous pericarditis whose treatment included: antituberculous medications with or without steroids and with variable rates of evacuation of the pericardium by either surgical or percutaneous means. Outcomes included recurrent pericardial effusion, recurrent tamponade, constrictive pericarditis requiring surgery and death.
With regards to mortality, in those studies with the shortest follow-up (6 months), the mortality ranged from 18% in a single-centre South African experience34 to 27.6% in the multinational IMPI registry.35 A longer follow-up (12 months) did not appear to change the mortality rates that remained between 17%25 and 23.8%.36 The study with the longest follow-up period reported an all-cause mortality rate of 25% at 10 years.36 Cumulatively, the information from these studies suggests that while mortality related to tuberculous pericarditis is high, most of it occurred early with good survival in those still alive at 6 months.
The evidence from around the continent suggests that survival remains poor in those with a diagnosis other than tuberculosis. Mortality among Zimbabwean patients with predominantly purulent and tuberculous pericarditis between 1967 and 1971 was 26.3%. In Togo, the mortality rate was 18% at 12 months13 and 30% in a single prospective South African hospital experience.25 In the aforementioned South African study, the fatality rate was up to 90% in patients with a malignant pericardial effusion.25 Survival with pericardial disease has been shown to be worse in HIV-infected patients compared with HIV-uninfected patients.35 39 For instance, clinical HIV disease doubled the risk of mortality from tuberculous pericarditis in the multinational IMPI study,35 and the mortality rate was up to 36.7% among persons living with HIV/AIDS with effusive pericardial disease from Mali.16
Seven studies reported on the outcome of surgical treatment of either constrictive pericarditis of large effusive pericarditis.24 26 29 38 40–42 Two studies reported on the outcome of pericardiectomy in large cohorts of patients with constrictive pericarditis. Both showed a consistent mortality rate of 14% in a group of patients in South Africa26 and 12.5% in a study in Ivory Coast.38 The other smaller studies were case series that showed a good clinical benefit and low perioperative complications rate of pericardiectomy or pericardiostomy for constrictive pericarditis or large pericardial effusions.24 29 40–42
This scoping review is the first to systematically summarise data on the prevalence, aetiology and outcomes of pericardial disease in Africa. The main findings from the review include: (1) a wide variation in the prevalence of pericardial disease according to the study population, from as low as 1.1% among patients presenting with cardiac symptoms14 to as high as 46.5% among those infected with HIV and with cardiac symptoms17; (2) tuberculosis was the most common cause of pericardial disease with HIV infection being the major risk factor. In a significant proportion of patients, no cause is ever found and in the remaining few, malignancies and other systemic diseases such as SLE may be present; and (3) although pericardial disease is a relatively infrequent cause of cardiac disease on the continent, it is important because of its very high 6–12 months mortality rate.
Approximately, one in four affected patients with pericarditis dies despite seemingly appropriate treatment. The highest mortality is observed in patients with malignant pericardial effusion (up to 90%), but this likely reflects the poor overall prognosis of malignancies and unavailability of effective treatment in resource-poor African settings.25 The mortality of tuberculous pericarditis was found to be 17% in HIV-uninfected individuals as compared with 40% in their HIV-infected counterparts.35 Potential explanations for the impact of HIV on mortality include a higher and more active mycobacterial load and a relatively higher prevalence of complications such as myopericarditis.48 49
As aforementioned, tuberculosis is the most common cause of pericarditis in Africa. The increasing incidence of tuberculous pericarditis over the past two decades has been driven by the HIV epidemic.3 In contrast to developed countries where the burden of HIV is low, tuberculosis is the cause in less than 4% of cases of pericarditis,43 and pericarditis is mostly idiopathic (in over 75% of cases) or due to neoplasms or autoimmune diseases.44 The aetiological diagnosis of pericardial diseases remain a challenge in Africa, especially when tuberculosis is suspected. Our review suggests that the cause of effusive and constrictive pericarditis is unknown in between 20% and 50% of patients. The low availability and affordability of microbiological tests and other relevant investigations in resource-poor settings make an accurate diagnosis very challenging. This is illustrated in the IMPI registry34 35 that reported that across the continent, the diagnosis of tuberculous pericarditis was made at the clinical discretion of the attending physician in the majority of cases34 with few centres adopting a standardised diagnostic approach to establishing a definitive diagnosis. In addition, the performances of culture-based and smear-based tests for tuberculous pericarditis are compromised due to a paucibacillary pericardial fluid (even though it may be multibacillary in advanced HIV48) in a greater majority of the patients,50 while the accuracy of histological diagnostic are reduced in the context of HIV infection as fewer granulomas are observed in those with severe immunosuppression.51 With the high morbidity and mortality associated with tuberculous pericarditis, especially in persons living with HIV/AIDS, current efforts towards the development of a highly sensitive and/or specific test are solicited. The affordability of new diagnostic tests to the average African is crucial for a greater impact.
Issues have been raised regarding the efficacy of the current regimen for tuberculous pericarditis. For instance, a study from the IMPI registry investigated the role of several microbial and clinical factors as predictors of mortality in patients with microbiologically proven tuberculous pericarditis.48 The study found that patients with culture positive tuberculous pericarditis had a high bacillary load in the pericardial fluid and that this bacillary burden was a significant predictor for both early and late mortality. The mortality was high, mostly due to failure of antituberculous treatment in patients on directly observed therapy, thus with good adherence. The high bacillary burden coupled with inadequate antituberculous drugs concentrations in the pericardial fluid due to poor penetration52 53 suggests the need to improve the efficacy of antituberculous treatment for tuberculous pericarditis by the design of a highly bactericidal regimen.48
This review is mainly limited by the predominance of studies with small sample sizes and a high degree of heterogeneity among the different study populations. This highlights the need for well-designed studies to better capture the burden and characterise the epidemiology of pericardial diseases in African populations.
Pericardial diseases represent a significant cause of morbidity and mortality in Africa, especially in the HIV-infected population. The aetiological diagnosis of pericardial diseases remain challenging in African resource-limited settings, with many cases being of unknown cause. Tuberculosis is the most frequent cause of pericardial diseases, and it is associated with poor outcomes. Associated myocarditis and high bacillary loads in the pericardial fluid contribute to the high mortality of tuberculous pericarditis.
A multinational African registry is needed to better characterise the epidemiology of pericardial diseases in African populations. As tuberculous pericarditis carries a high burden in these populations, especially in the context of the HIV epidemics, reliable diagnostic tests should be developed for early diagnosis of the disease and timely treatment. Research efforts should be directed towards designing an anti-tuberculous regimen with high pericardial permeability and which is highly bactericidal in order to improve the outcome of tuberculous pericarditis.
ALN, JRN, AK and UFN contributed equally.
Contributors JJN conceived the study. JJN did the literature searched and selected studies with VNA. JJN, UFN, ALN, AK, JRN and VNA collected data. JJN summarised and interpreted the data. JJN, VNA and MN drafted the manuscript. All authors revised the manuscript for intellectual content. All authors approved the final version of the manuscript. JJN is the guarantor of the review.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent Not required.
Provenance and peer review Not commissioned; externally peer reviewed.