2. Cardiovascular Involvement in TB
The cardiovascular structures most involved in tuberculosis are the pericardium, myocardium, and aorta (
Figure 1).
Figure 1. Cardiovascular involvement in TB.
2.1. Tuberculous Myocarditis
Myocardial involvement may be associated with pericarditis in the form of myopericarditis, or it may cover other clinical scenarios. Tuberculous myocarditis is particularly rare, with an estimated prevalence of less than 2%
[34,35][16][17].
The first two cases of myocardial damage in TB were reported in 1664 by Maurocadat and in 1761 by Morgagni
[36][18]. Epidemiological studies show a predominance of tuberculous myocarditis in patients under 45 years old, and it is twice as common in male patients
[15]. Invasion of the myocardium by MTB is realized via the hematological route, by retrograde lymphatic insemination from the mediastinal nodes, or by direct invasion from the pericardium
[37,38,39][19][20][21]. The different effects of MTB on the pericardium and myocardium can be explained on the one hand by the continuous movement of the myocardium, which indirectly prevents the lodging of bacilli, and on the other hand by the lactic acid produced, which has a protective role against bacilli
[36,40][18][22].
Most cases have been reported in immunocompromised patients, who frequently have an HIV infection. Reported cases have frequently affected the left heart, and especially the left ventricle. Predominant right mediastinal lymph node involvement has been observed in many patients with MTB-induced myocarditis, which increases the risk of right heart damage by contiguity
[41][23].
Myocardial damage is frequently asymptomatic, sometimes with severe consequences, leading to forms of acute heart failure
[42,43][24][25]. In these cases, involvement of the right mediastinal lymph nodes has been observed, with a greater chance of contiguous involvement of the right side of the heart
[15]. Identified forms include nodular myocardial damage with central caseation, miliary forms, or diffuse, inflammatory, giant cell forms
[44][26].
A significant percentage of patients with myocardial damage also have pericarditis, which further worsens the prognosis
[15]. Studies presenting data from endomyocardial biopsies indicate predominantly biventricular involvement in about 70% of cases, with isolated right ventricular dysfunction occurring in a small percentage of cases, only 8%
[45][27].
The symptomatology of patients with myocarditis secondary to TB is variable, ranging from no symptoms to severe forms presenting at onset, with ventricular arrhythmias
[46][28], sudden death, long QT syndrome
[47][29], atrioventricular blocks, or clinical signs of congestive heart failure
[42,48,49][24][30][31]. The clinical picture of these patients includes in some cases electrolyte imbalances, one of the most frequently reported being hypercalcemia
[50][32].
The diagnostic criteria for myocarditis are the classic ones, represented by the identification of a high titer of myocardial enzymes together with the echocardiographic presence of left ventricular systolic dysfunction
[15,37,51][15][19][33]. Nuclear magnetic resonance is an essential investigation in patients with myocarditis, highlighting in the T2 sequence a central and peripheral hypointense signal as well as a hyperintense thin line
[52,53,54,55][34][35][36][37].
There are few data reported in the literature on the treatment of patients with tuberculous myocarditis, with clinical trials recommending the initiation of etiologic treatment. Improvement in symptoms does not eliminate the associated risk of sudden death, so these patients require regular monitoring, most often by multidisciplinary teams
[56,57][38][39].
The most common complications reported were atrial fibrillation and sudden cardiac death
[58][40]. Fulminant forms of MTB myocarditis can have an unfavorable outcome, especially in immunocompromised patients. Clinical studies report that 80% of fatal cases occur in female patients with associated LV systolic dysfunction
[15].
Cases of chronic heart failure with preserved systolic function due to extensive intra-myocardial calcifications associated with latent MTB infection have also been reported
[59][41].
2.2. Coronary Artery Disease and Tuberculosis
One interaction being considered is that between TB and coronary atherosclerosis, the presence of TB being associated with a 1.76-fold increased risk of developing coronary artery disease
[27,60][42][43]. Implicitly, patients with TB have an associated risk of acute myocardial infarction of 1.98 compared to a similar cohort of patients without TB
[23,61,62][44][45][46].
Both TB and ischemic coronary artery disease are common in developing countries and their association is all the more frequent. Several mechanisms are thought to be behind this. A first mechanism is related to a chronic inflammatory reaction, cell-mediated immune activation with the release of cytokines and chemokines, following latent infection. A second mechanism is the initiation of an autoimmune process following chronic infection, with production of antibodies against mycobacterial heat shock protein-65 (HSP65)
[63][47]. This causes an induced cross-reaction with human HSP65, leading to endothelial injury and stimulating atherogenesis.
Heat shock proteins are a homogeneous group of proteins that arise in response to stress factors, originally discovered as a reaction to heat, hence the name. They show a high homogeneity between species and have in particular a chaperone role, but also mediate immune reactivity in certain diseases
[64][48]. Animal model studies have shown that HSP65 inhibition affects IL-10 and paraoxonase-1 activity, while interferon-γ expression, myeloperoxidase activity, and the high-density lipoprotein inflammatory index tend to increase, leading to generalized as well as aortic atherosclerosis
[65][49]. It has been observed that latent infection also results in elevated levels of interferon-γ, which may be a good predictor of progression to clinically manifest disease
[66,67][50][51]. Moreover, considering that HSP65 mediates the early stages of the atherogenesis process, it is also being studied for the development of an anti-atherosclerotic vaccine
[68][52]. A population-based study of more than 10,000 patients showed a 1.4-fold increased risk of acute coronary syndrome in patients diagnosed with TB compared to the general population
[69][53]. This effect may be related to a combination of factors, with lung inflammation in general presumed to induce systemic inflammatory response, endothelial dysfunction, and atheroma plaque destabilization
[70][54].
Recent data on latent TB have shown a high prevalence of ischemic coronary artery disease among these patients. Even in the absence of clinically manifest TB, chronic immune response to MBT can intensify the atherosclerotic process
[15]. Another recent study showed a twofold increased likelihood of association of latent TB with acute myocardial infarction, after correcting for classical cardiovascular risk factors and other confounders
[61][45]. Furthermore, it appears that vascular damage is not limited to the coronary arteries, and an association of latent TB with both peripheral arterial disease and ischemic stroke has been observed
[71,72][55][56].
C-reactive protein (CRP), total white blood cell count, and neutrophil-to-lymphocyte ratio are three independent inflammatory predictors associated with a negative prognosis in coronary artery disease
[73,74][57][58]. Serum CRP correlates with MTB bacterial load in sputum, having prognostic value and being associated with a high risk of death
[75,76,77][59][60][61].
The beneficial effect of statins in reducing associated cardiovascular risk and decreasing systemic inflammation has been demonstrated, but its modulatory role in combination with MTB has not been fully elucidated to date
[78,79,80,81,82][62][63][64][65][66]. There is a causal relationship between cholesterol and MTB, the bacterial agent needing cholesterol for infection and survival, with the caveat that the progression of infection is correlated with the ability of the immune system to limit infection
[83,84][67][68]. Oxidized low-density lipoprotein plays an important role in patients with type 2 diabetes mellitus and TB, playing a central role in the formation of lipid-loaded foamy macrophages that contribute to the progression of tuberculous granulomas through lysosomal dysfunction
[85,86,87][69][70][71].
Thus, various preclinical studies are reported in the literature in which statin administration is associated with stimulation of autophagy and phagosome maturation of MTB-infected macrophages. Research in murine models also highlights the beneficial role of statin administration in enhancing the therapeutic effect of first-line antituberculous drugs
[88,89,90,91][72][73][74][75].
Statins have a beneficial role in the treatment of patients with TB infection and can be used as an adjuvant medication to standard treatment
[83,92,93,94][67][76][77][78]. Administration of this hyperlipidemic medication increases cell resistance to MTB, but further clinical trials are needed in this research direction
[92,95,96,97,98][76][79][80][81][82].
2.3. Tuberculous Pericarditis
Tuberculous etiology of pericarditis is one of the most common along with the neoplastic etiology
[99][83], with a prevalence depending on a country’s level of development
[100][84]. Approximately 1–2% of TB patients have associated pericarditis
[101][85]. This type of pericarditis is characterized by a significant inflammatory status
[102][86], chronicity
[101][85], and a high risk of progression to a constrictive form
[103][87].
Pericardial insemination with MTB occurs retrogradely, by the lymphatic route, by hematological dissemination, or in rare cases by direct damage to surrounding structures such as the lungs, pleura, or spine
[104][88]. In the case of HIV co-infection, the pathway of dissemination is hematological
[105][89]. The most common form of presentation of tuberculous pericarditis is the effusive form (in about 80% of cases), the constrictive form being considered one of the most common sequelae
[106][90].
The prevalence of the constrictive form of pericarditis is reduced in patients without TB
[100][84]. In patients infected with MTB, this form of the disease is seen in 25% of cases, but this percentage may be higher than for other forms such as idiopathic or viral
[107][91].
Some patients have an atypical clinical form that poses problems of diagnosis and treatment, often delaying the latter and thus worsening the prognosis of patients. Tuberculous pericarditis presents four distinct stages, with a clinical picture and imaging features specific to pathophysiological processes (
Figure 2)
[13].
Figure 2. Stages of constrictive pericarditis.
The dry stage is the least common, despite the marked symptoms that accompany it. The effusive stage is most often diagnosed by echocardiography, with a corresponding clinical worsening of the patient’s general status through the appearance of heart failure or even cardiac tamponade. The constrictive stage is encountered in a variable percentage of patients ranging from 5–25%, representing the stage with the most reserved prognosis in terms of associated dysfunction.
Multimodal imaging evaluation of patients with pericardial effusion includes echocardiography, computed tomography, and nuclear magnetic resonance to differentiate the constrictive form from restrictive cardiomyopathy, which is the main entity with which the differential diagnosis is made
[108,109,110][92][93][94] (
Figure 3).
Figure 3. Multimodal imaging evaluation in tuberculous pericarditis.
Karima et al.
[112][95] analyzed a group of 25 patients with constrictive pericarditis and demonstrated a high prevalence of infectious etiology as well as a statistically significant association with the presence of right ventricular dysfunction in this category of patients. Echocardiography, computed tomography, or nuclear magnetic resonance are the main imaging investigations used in the diagnostic algorithm of pericardial effusion
[113][96]. The main findings identified in constrictive pericarditis as well as echocardiographic arguments establishing the differential diagnosis with restrictive cardiomyopathy are shown in
Figure 4.
Figure 4. Multimodal imaging evaluation of patients with pericardial effusion, constrictive pericarditis, and restrictive cardiomyopathy.
If transthoracic echocardiography provides suboptimal or inconclusive images (especially in the context of high suspicion of cardiac tamponade), transesophageal echocardiography is recommended. Cardiac CT is the gold standard in the evaluation of pericardial calcifications. Contrast-enhanced CT is recommended to avoid overestimation of pericardial effusion and to prevent artefacts from being missed on native examination. Nodular areas with increased attenuation, calcification of the anterior pericardium, and lack of changes when changing position in decubitus or in the presence of contrast enhancement of pericardium are arguments in favor of pericardial thickening
[99,114,115][83][97][98].
Given the fact that a high percentage of patients have concomitant pulmonary and extra-pulmonary involvement, the management of these patients must be integrative, focused on the use of imaging methods. In addition to the above, chest ultrasound is another assessment method with applications in the management of patients with chest TB. Chest ultrasound allows detection of TB, dynamic follow-up of pleural effusions after evacuation, biopsy, or assessment of nodular involvement in children
[116,117,118][99][100][101].
Geographical location often guides the diagnosis of a pericardial effusion, and there are a number of arguments for a tubercular etiology in endemic countries
[119][102]. Thus, the identification of MTB in the stained smear or pericardial fluid culture and the presence of granulomas on histopathological examination confirm the diagnosis of tuberculous pericarditis
[120][103]. The presence of pericarditis in a patient diagnosed with TB, an increased adenosine deaminase activity (ADA) activity and a high percentage of lymphocytes in the pericardial fluid, and a favorable clinical response secondary to the initiation of antituberculous treatment provide diagnostic clues, but further clinical tests are required to establish a positive diagnosis
[106,115][90][98].
In some particular situations, obtaining negative serological tests does not exclude the tuberculous etiology of a pericardial effusion, sometimes requiring biopsy
[16][104]. Patients with constrictive tuberculous pericardial disease have elevated levels of pericardial inflammatory cytokines such as IL-10 (
p = 0.006) and interferon-gamma (
p = 0.03). Ntsekhe et al.
[121][105] analyzed a cohort of 91 patients with constrictive pericarditis, 68 of whom had TB, and using statistical regression analysis identified right atrial pressure above 15 mmHg (odds ratio of 48,
p < 0.001) and serum IL-10 levels above 200 pg/ml (
p = 0.04, 10 times higher associated risk) as predictors associated with calcification of the pericardium. In addition to inflammatory cytokine changes, anemia is one of the most common hematological changes seen in TB patients. The incidence reported in the literature varies, from 32% to 94%, most commonly in normochromic, normocytic forms
[122,123][106][107].
De Vita et al.
[124][108] reported the case of a 21-year-old patient with TB pericarditis as the first manifestation in whom pericardial fluid analysis was negative for MTB infection. Positive diagnosis for TB was established by positive urine lateral flow lipoarabinomannan assay, which required initiation of antituberculosis treatment. The glycolipid lipoarabinomannan is released during metabolism and degradation of infected cells in patients with active TB, leading to downregulation of pathophysiological processes that reduce interferon-gamma and interleukin-12 secretion
[125,126][109][110]. Since its first use in 2001 to date, several clinical trials have been conducted to test the efficacy of this diagnostic test, but the reported results show low sensitivity
[127,128][111][112].
The concurrent presence of HIV infection negatively modulates the prognosis of patients with tuberculous pericarditis, with an associated twofold increased risk of death
[129][113]. In patients with tuberculous pericarditis without HIV infection, the virulence of the pathogen and the immune response secondary to interactions in the pericardium correlate with the course of the disease and thus with the patients’ medium- and long-term prognosis
[130,131][114][115]. In HIV-positive patients, pericardial involvement frequently occurs through dissemination, in which case the infection plays the main prognostic role
[105][89].