Frequency of pleural effusion in dengue patients by severity, age and imaging modality: a systematic review and meta-analysis

BACKGROUND

Identification of pleural effusion (PE) in dengue infection is an objective measure of plasma leakage and may predict disease progression. However, no studies have systematically assessed the frequency of PE in patients with dengue, and whether this differs across age and imaging modality.

METHODS

We searched Pubmed, Embase Web of Science and Lilacs (period 1900-2021) for studies reporting on PE in dengue patients (hospitalized and outpatient). We defined PE as fluid in the thoracic cavity detected by any imaging test. The study was registered in PROSPERO (CRD42021228862). Complicated dengue was defined as hemorrhagic fever, dengue shock syndrome or severe dengue.

RESULTS

The search identified 2,157 studies of which 85 studies were eligible for inclusion. The studies (n = 31 children, n = 10 adults, n = 44 mixed age) involved 12,800 patients (30% complicated dengue). The overall frequency of PE was 33% [95%CI: 29 to 37%] and the rate of PE increased significantly with disease severity (P = 0.001) such that in complicated vs. uncomplicated dengue the frequencies were 48% and 17% (P < 0.001). When assessing all studies, PE occurred significantly more often in children compared to adults (43% vs. 13%, P = 0.002) and lung ultrasound more frequently detected PE than conventional chest X-ray (P = 0.023).

CONCLUSIONS

We found that 1/3 of dengue patients presented with PE and the frequency increased with severity and younger age. Importantly, lung ultrasound demonstrated the highest rate of detection. Our findings suggest that PE is a relatively common finding in dengue and that bedside imaging tools, such as lung ultrasound, potentially may enhance detection.

Potential role of transthoracic echocardiography for screening LV systolic dysfunction in patients with a history of dengue infection. A cross-sectional and cohort study and review of the literature

Background

Dengue virus can affect the cardiovascular system and men may be at higher risk of severe complications than women. We hypothesized that clinical dengue virus (DENV) infection could induce myocardial alterations of the left ventricle (LV) and that these changes could be detected by transthoracic echocardiography.

Methodology/Principal findings

We examined individuals from Acre in the Amazon Basin of Brazil in 2020 as part of the Malaria Heart Study. By questionnaires we collected information on self-reported prior dengue infection. All individuals underwent transthoracic echocardiography, analysis of left ventricular ejection fraction (LVEF) and global longitudinal strain (GLS). We included 521 persons (mean age 40±15 years, 39% men, 50% urban areas) of which 253 (49%) had a history of dengue infection. In multivariable models adjusted for clinical and sociodemographic data, a history of self-reported dengue was significantly associated with lower LVEF (β = -2.37, P < 0.01) and lower GLS (β = 1.08, P < 0.01) in men, whereas no significant associations were found in women (P > 0.05). In line with these findings, men with a history of dengue had higher rates of LV systolic dysfunction (LVEF < 50% = 20%; GLS < 16% = 17%) than those without a history of dengue (LVEF < 50% = 7%; GLS < 16% = 8%; P < 0.01 and 0.06, respectively).

Conclusions/Significance

The findings of this study suggest that a clinical infection by dengue virus could induce myocardial alterations, mainly in men and in the LV, which could be detected by conventional transthoracic echocardiography. Hence, these results highlight a potential role of echocardiography for screening LV dysfunction in participants with a history of dengue infection. Further larger studies are warranted to validate the findings of this study.

Is self-rated health associated with cardiovascular risk factors and disease in a low-income setting? A cross-sectional study from the Amazon Basin of Brazil

OBJECTIVE

Prior studies have suggested that self-rated health may be a useful indicator of cardiovascular disease. Consequently, we aimed to assess the relationship between self-rated health, cardiovascular risk factors and subclinical cardiac disease in the Amazon Basin.

DESIGN

Cross-sectional study.

SETTING, PARTICIPANTS AND INTERVENTIONS

In participants from the Amazon Basin of Brazil we obtained self-rated health according to a Visual Analogue Scale, ranging from 0 (poor) to 100 (excellent). We performed questionnaires, physical examination and echocardiography. Logistic and linear regression models were applied to assess self-rated health, cardiac risk factors and cardiac disease by echocardiography. Multivariable models were mutually adjusted for other cardiovascular risk factors, clinical and socioeconomic data, and known cardiac disease.

OUTCOME MEASURES

Cardiovascular risk factors and subclincial cardiac disease by echocardiography.

RESULTS

A total of 574 participants (mean age 41 years, 61% female) provided information on self-rated health (mean 75±21 (IQR 60-90) points). Self-rated health (per 10-point increase) was negatively associated with hypertension (OR 0.87 (95% CI 0.78 to 0.97), p=0.01), hypercholesterolaemia (OR 0.89 (95%CI 0.80 to 0.99), p=0.04) and positively with healthy diet (OR 1.13 (95%CI 1.04 to 1.24), p=0.004). Sex modified these associations (p-interaction <0.05) such that higher self-rated health was associated with healthy diet and physical activity in men, and lower odds of hypertension and hypercholesterolaemia in women. No relationship was found with left ventricular ejection fraction <45% (OR 0.97 (95% CI 0.77 to 1.23), p=0.8), left ventricular hypertrophy (OR 0.97 (95% CI 0.76 to 1.24), p=0.81) or diastolic dysfunction (OR 1.09 (95% CI 0.85 to 1.40), p=0.51).

CONCLUSION

Self-rated health was positively associated with health parameters in the Amazon Basin, but not with subclinical cardiac disease by echocardiography. Our findings are of hypothesis generating nature and future studies should aim to determine whether assessment of self-rated health may be useful for screening related to policy-making or lifestyle interventions.

TRIAL REGISTRATION NUMBER

Clinicaltrials.gov: NCT04445103; Post-results.