There was a high prevalence of echocardiographic abnormalities found in this study of HIV infected individuals. The prevailing abnormalities identified were diastolic dysfunction, regional wall motion abnormality, LVH and left atrial dilation.
Diastolic dysfunction (DD), predominantly grade I (impaired relaxation) was reported in 30.1% of the participants with normal systolic function. Similar to our finding, diastolic dysfunction has been reported to be the most common finding previously (7, 9, 18, 19). High prevalence ranging from 42.8% in India to as high as in 85.7% in Congo have been reported in the past (3, 7, 9, 18, 19). The lowest report of diastolic dysfunction was from a study on patients with early HIV infection of only 2% (20). Other studies report prevalence from 9.25% to 19.9% (2, 21) with a recent systematic review reporting pooled prevalence of 29.3%, a number similar to our study (4). Similar to this study, increasing age has been shown to be an independent risk factor for DD in HIV patients (9) and is also a known risk factor in the general population (22). Hypertension and LVH are other risks for DD seen in previous studies also supported by our findings (3, 5, 9, 22, 23). HIV infection is also a risk factor for DD by itself (3). The level of immunosuppression and viral load did not affect finding of DD in this or some other studies (3, 7). The mechanisms behind DD in HIV infection is multifactorial with hypertension, LVH, direct myocardial viral infection (HIV itself, Mycobacterium avium, Cryptococcus neoformans, Toxoplasma gondii), neoplastic involvement (Lymphoma, Kaposi Sacroma), infiltrative diseases (amyloidosis, tuberculosis), increased inflammation, immunosuppression, increased HIV related atherosclerosis, Nucleoside/Nucleotide reverse transcriptase inhibitor (NRTI) induced mitochondrial damage and cardiomyopathy all potentially contributing (3, 18). DD precedes systolic dysfunction and is considered an early finding of myocardial disease that needs follow up (19).
LVH was seen in 10.3% of our study participants. Previous studies have reported prevalence of LVH from 4% to 46.9% (2, 3, 18, 23-25). This variation could be explained by the use of different methods for assessment of LVH including indexing for body surface area and height which was not done in our study. A study done on patients receiving ART showed hypertension to be a risk for LVH, supporting a similar finding in our study (23). Age and hypertension are known risk factors for LVH in the general population (26). The severity of HIV infection was not shown to be a risk for LVH which was also the case from previous findings (3, 23-25). A study that compared HIV infected patients to controls have shown that the infection itself increases risk of LVH (3). Several factors contribute to LVH in HIV. Infection of myocardium and subsequent local release of cytokines, opportunistic infections, use of ART particularly protease inhibitors, subclinical atherosclerosis and state of increased inflammation may all play a role (3, 24). LVH is an important precursor to the development of hypertensive heart disease and is associated with cardiovascular mortality and therefore needs close follow-up and treatment (9, 26).
Diagnosis of ischemic heart disease was made after the echocardiographic finding of pathologic contraction (regional wall motion abnormality) was seen in 22.2%. This is higher than most previous studies which reported regional wall motion abnormality from 1 % to 23.2% (7, 9, 10, 21, 27). This large difference could be explained by the exclusion of patients with known ischemic heart disease except in two of the studies which had the higher prevalence of 13.5% and 23.2% which is close to our finding (9, 10). In addition to HIV infection itself, traditional cardiovascular risk factors such as increasing age, male gender and diabetes are supplementary risk factors (28, 29). HIV infected patients have increased risk for coronary artery disease with increasing reports (7, 8). The major proposed mechanism is increased endothelial dysfunction but also includes lipodystrophy and impaired glucose metabolism (7, 8). Certain ART medications increase level of dyslipidemia, insulin resistance and C-reactive protein that increase the risk of coronary artery disease (5, 8). They also have increased prevalence of traditional cardiovascular risk factors (5). With increasing longevity and decreasing opportunistic infections, coronary artery disease is expected to become an increasing burden (8). Regional wall motion abnormality is an earlier stage of global LV dysfunction that mandates appropriate management (7).
Left atrial enlargement (LAE) was seen in 8.1% of our participants. Previous studies reported atrial enlargement from 2.4%, 7.7% and 40% (2, 9, 23). The difference could be explained by the use of different cutoff points; usage of linear measurement of LA anteroposterior size and not LA volume measurement in our study; and lack of adjustment made for body surface area. Measurement of linear dimension as opposed to volume does not reflect accurate LA size as the LA enlarges in different dimensions during LA remodeling (13). A study comparing HIV infected and non-infected controls showed larger LA size in infected individuals (3). Hypertension was the only factor independently associated with LAE in our study. Previous studies also found associations with hypertension and LVH (2, 23). Similar to our finding, CD4 and viral load have not found to be significant previously (23). LAE in the general population, has been shown to be a risk factor for cardiovascular disease including heart failure, atrial fibrillation and stroke (23, 30).
Pericardial effusion was seen in only 6 (2.1%) of the participants which was mostly small or moderate and none of the patients had cardiac tamponade that needed drainage. There is a trend toward decreasing pericardial effusion in HIV patients particularly in developed countries (5). Studies from Congo and Cameroon showed prevalence of effusion of 20.4% and 46.5% respectively, showing very high prevalence in Africa (31, 32). Others reported prevalence of as low as 4% to 16.6% (7, 8, 10, 21, 27, 30, 33, 34) and in one study only two patients had pericardial effusion (35). There is a difference of the participants of the studies with some including patients with profound immunosuppression. Pericardial effusion is seen mainly in patients with lower CD4 counts (10, 21, 27, 33). The pathophysiology includes pericarditis caused by the virus and opportunistic infections such as Mycobacterium Tuberculosis, Cryptococcus Neoformans and Herpes Simplex (30). Malignancies such as Lymphoma and Kaposi Sarcoma which have become less common with the widespread use of ART are other potential causes (5, 30). Patients with pericardial effusion potentially have higher mortality and therefore needs management as many of the causes are potentially treatable (31).
Reduced ejection fraction was seen in only one patient. A systematic review and meta-analysis published in 2018 showed the pooled prevalence of reduced ejection fraction below 50% was 12.3% which was more reduced in patients with AIDS showing the possible relation with immunodeficiency and opportunistic infections and showed lower prevalence in patients on ART potentially explaining the preserved ejection fraction in our patients (4). There are some studies where four percent or less of the studied population had low ejection fraction (2, 3, 20).
Similarly, the systematic review and meta-analysis reported prevalence of pulmonary hypertension in 11.5% of patients while this study revealed prevalence of 3.6% (4). A study done in similar Ethiopian setting showed prevalence of 14% (16). However, other studies also revealed low prevalence of 2.6% and 3.3% (2, 7). The sensitivity of echocardiography for pulmonary hypertension is not satisfactory and invasive tests may be required in patients with strong clinical suspicion as it is associated with increased mortality (16, 36).
There are certain limitations to this study. The cross-sectional nature of the study hinders from making final conclusions about echocardiographic findings and associated factors. There was also no control group for comparison. Despite these limitations our study provides a detailed assessment of echocardiographic findings in a large sample of HIV infected patients and provides new information for the population in the country that was limited to only few echocardiographic parameters. We recommend for future studies to be done taking our limitations into consideration and using a longitudinal study nature that takes the effect of time and long ART use into account.