nueva página del texto (beta)
Artículo en XML
Referencias del artículo
Traducción automática
Enviar artículo por email
Citado por SciELO 
Similares en
SciELO 
Permalink
Introduction
Pericardial effusion (PE) is the abnormal accumulation of fluid in the space between the two layers of the pericardium. It is an increasingly common condition, due to the advance in diagnostic images that allow its identification.1
Multiple diseases can cause pericardial effusion, however, only in a minority of cases an accurate diagnosis can be made. Recent advances in diagnostic techniques (microbiology, for example) and newer imaging methods, have allowed to establish with greater precision its etiology.2
Etiologies imply great variations according to regions, they are not the same in developed countries as in developing countries. In the former, the majority are considered idiopathic or secondary to cancer, whereas in developing countries infectious etiologies, especially tuberculosis, are the main causes and represent more than a half of the cases.3 Knowing the local epidemiology is essential for clinical practice, physicians will be more confident regarding which etiologies to asses when facing a severe pericardial effusion and making the correct decision between diagnostic and treatment alternatives. Likewise, it is worth asking if the impact of PE secondary to tuberculosis is as great as it is presumed in affected countries or if there are other etiologies with relevant participation.
Prognosis associated with PE depends on the underlying etiology, however, having PE is a marker of severity and in some cases, leading to an ominous outcome.4-6 PE, specifically in patients with human immunodeficiency virus (HIV) is associated with low survival, 36% at 6 months and 19% at 1 year; due to the antiretroviral therapy and the timely diagnosis and treatment of the effusion, these statistics have been reduced and the paradigm of the disease has changed.7
Although in cases of mild PE, a specific treatment is not recommended other than treating the root cause, adequate clinical and imaging follow-up is essential to prevent progression to severe effusion and cardiac tamponade; the most feared complication and with the highest mortality within the spectrum of the disease.8,9
There is no history of local studies that establish the different etiologies of severe PE in Colombia. The aim of this study is to describe the main etiologies and comorbidities at a tertiary care hospital in the city of Medellín.
Material and methods
Retrospective case series, based on clinical records obtained between November 1st, 2006 and December 31st, 2018 from the emergency and hospitalization service of a highly complex hospital in the city of Medellín, where patients are mostly from the public health sector and rural areas. This hospital has the services of internal medicine, cardiology, infectious diseases, general surgery, cardiovascular surgery and intensive care unit.
The included patients were adults older than 18 years who entered the emergency department or hospitalization. All patients had an imaging diagnosis of PE and, its severity was defined by echocardiographic quantification performed by an echocardiography cardiologist. The exclusion criteria were the absence of more than 50% of the data in the clinical history and the recurrence of the pericardial effusion after its first drainage. All the patients required some intervention for pericardial fluid drainage; percutaneously in 26 patients and with surgery (pericardial window) other 22.
Variables included were patient past medical history, drainage indication, type of drainage intervention and etiology of the PE, if successfully established. Etiologies were classified in two large groups: infectious or non-infectious, with subsequent specific definition by subgroups, according to the 2015 ESC Guidelines for the diagnosis and management of pericardial diseases.10
Descriptive analysis was performed with relative and absolute frequencies for the variables studied with the Stata version 12.1 software.
Results
During the study period, 2,553 patients were evaluated for pericardial effusion, 350 compatible with moderate to severe PE; 302 of these patients did not require percutaneous or surgical intervention. The clinical histories of 48 patients with severe pericardial effusion were identified in whom echocardiography and drainage of the pericardial effusion was performed (Figure 1). Of these patients, 50% were men, mean age was 52 ± 17.5 years. The most relevant comorbidities were hypertension (41%), malignancy (31%), chronic kidney disease (20%) and heart failure (16%). Other less frequent were diabetes mellitus, hypothyroidism, HIV infection, tuberculosis, and drug dependence (Table 1).
Table 1: Demographic and clinical characteristics.
| Characteristics | Measure % | Number of patients |
|---|---|---|
| Men | 50.00 | 24 |
| Age | 52.41 ± 17.5 years | 48 |
| Comorbidities | ||
| Hypertension | 41.00 | 19 |
| Cancer | 31.00 | 14 |
| Chronic kidney disease | 20.00 | 9 |
| Heart failure | 16.00 | 8 |
| Type 2 diabetes mellitus | 14.58 | 7 |
| Chronic obstructive pulmonary disease | 12.50 | 6 |
| Hypothyroidism | 8.33 | 4 |
| Autoimmune disease | 6.25 | 3 |
| Tuberculosis | 6.25 | 3 |
| HIV | 4.17 | 2 |
| Hyperthyroidism | 4.17 | 2 |
| Substance abuse | 2.08 | 1 |
HIV = human immunodeficiency virus.
The main indication for drainage was the need to find an etiology (58%), followed by hemodynamic compromise (23%) and symptoms refractory to medical treatment (19%).
Regarding the etiologies (Figure 2), non-infectious etiologies were the most frequent (66.7%). These included: malignancy (14 cases equivalent to 43.8% of non-infectious causes), postoperative or traumatic (12 cases, 37.5%), chronic kidney disease (2 cases, 6.2%), heart failure (2 cases, 6.2%) and autoimmune disease (2 cases, 6.2%). In 20.8% of the cases, it was not possible to establish a clear underlying cause, therefore, they were established as idiopathic PE or idiopathic pericarditis. Infectious etiologies were the least common (6 cases, 12.5%). Among these, pericardial tuberculosis was the cause in 5 of the 6 cases (83.3%).
Discussion
In this study, we identified that non-infectious etiologies were the most frequent etiologies in patients with severe PE. A significant fraction of these cases were secondary to malignancy, even with a mean age of 50 years, and consequently, patients presented with manifestations of malignant PE. Among the infectious etiologies, the main one was tuberculosis, this is possibly explained because Colombia has an intermediate prevalence for this entity, along with the fact that the institution where this study was carried out is a highly complex center with a large flow of patients presenting with extrapulmonary manifestations of tuberculosis. Pericardial biopsy was required in most patients with pericardial tuberculosis, to optimize the diagnostic performance of the tests used.
According to the published series so far, it is evident that the etiology of PE, especially in moderate to severe cases, varies dramatically when analyzed between developed or developing countries. In the former, most are idiopathic (50%), followed by malignancy (10-25%), pericarditis or infectious (15-30%), iatrogenic (15-20%) and associated with connective tissue diseases (5-15%). In developing countries, more than 60% of the cases are infectious, represented in more than half of the cases by tuberculosis, especially in those regions where this mycobacterium is endemic. It should be noted that HIV infection continues to rise worldwide, increasing the incidence of PE diagnosis related to this condition.11-15
Regarding pericarditis with associated pericardial effusion, it is mostly represented by infectious and malignant etiologies with a global distribution of 15-50%, depending on the series reviewed.11-14 Idiopathic pericarditis, which is presumed to be mostly due to post-viral causes, is the main inflammatory cause of pericardial effusion.16 The difficulty in diagnosing PE, those considered idiopathic, may be due to the fact that isolating a virus is a complex and difficult task, often requiring a pericardium sample for histological, cytological and/or immunohystological analysis. In the vast majority of cases, clarification of the etiology is not necessary for the management of the patient; furthermore, it would increase costs for the health system and imply invasive and additional procedures or interventions for the patient.17
It is important to highlight how progress in the different diagnostic methods (microbiological cultures, polymerase chain reaction, cardiac magnetic resonance imaging) has favored the identification of the underlying cause, making idiopathic etiologies group, to decrease. In this study, idiopathic etiology was 20.8%, while in the world literature it is approximately 50%.10
On the other hand, the high ratio of severe pericardial effusion and cardiovascular surgeries should be kept in mind when the patient’s postoperative period does not show a favorable evolution. This high prevalence described in the study could be maximized by the fact that the patients were analyzed in a hospital with high-complexity of services such as thoracic and cardiovascular surgery.
Among the study limitations, it is a retrospective study and the research was carried out in a single center, which could disregard other PE etiologies. It should be noted that, in this setting, severe PE drainage procedures are only performed in highly complex hospitals and clinics. On the other hand, the amount of patient data described is not large despite the fact that the medical records reviewed, included more than 10 years, possibly explained by the fact that only patients with severe pericardial effusion with drainage were included, since patients who did not have a study of pericardial fluid were not included. Nevertheless, in Colombia there are no similar reports, as the presented in this study.
Conclusions
PE is an entity with an important prevalence and associated morbidity and mortality, often with insufficient resources aimed at finding its etiology. In this study, non-infectious causes were the most common, especially those related to traumatic or postsurgical events, making it easier to suspect, timely diagnose, and to intervene. However, infectious (especially pericardial tuberculosis) or idiopathic (possibly post-viral) causes also account for a significant number of cases, a situation that resembles with the reports in world literature. In this study, the idiopathic etiology was lower than that reported in other series, suggesting that an exhaustive and rigorous search has been carried out, which is essential to achieve an adequate diagnostic and therapeutic approach.
