key: cord-0745819-urcsqxtu
authors: Aggarwal, Ashutosh Nath; Agarwal, Ritesh; Dhooria, Sahajal; Prasad, Kuruswamy Thurai; Sehgal, Inderpaul Singh; Muthu, Valliappan
title: Active pulmonary tuberculosis and coronavirus disease 2019: A systematic review and meta-analysis
date: 2021-10-21
journal: PLoS One
DOI: 10.1371/journal.pone.0259006
sha: 6b0d12c9d53df209e4b03a9adf6d935b0ddbf137
doc_id: 745819
cord_uid: urcsqxtu

OBJECTIVE: The proportion of COVID-19 patients having active pulmonary tuberculosis, and its impact on COVID-19 related patient outcomes, is not clear. We conducted this systematic review to evaluate the proportion of patients with active pulmonary tuberculosis among COVID-19 patients, and to assess if comorbid pulmonary tuberculosis worsens clinical outcomes in these patients. METHODS: We queried the PubMed and Embase databases for studies providing data on (a) proportion of COVID-19 patients with active pulmonary tuberculosis or (b) severe disease, hospitalization, or mortality among COVID-19 patients with and without active pulmonary tuberculosis. We calculated the proportion of tuberculosis patients, and the relative risk (RR) for each reported outcome of interest. We used random-effects models to summarize our data. RESULTS: We retrieved 3,375 citations, and included 43 studies, in our review. The pooled estimate for proportion of active pulmonary tuberculosis was 1.07% (95% CI 0.81%-1.36%). COVID-19 patients with tuberculosis had a higher risk of mortality (summary RR 1.93, 95% CI 1.56–2.39, from 17 studies) and for severe COVID-19 disease (summary RR 1.46, 95% CI 1.05–2.02, from 20 studies), but not for hospitalization (summary RR 1.86, 95% CI 0.91–3.81, from four studies), as compared to COVID-19 patients without tuberculosis. CONCLUSION: Active pulmonary tuberculosis is relatively common among COVID-19 patients and increases the risk of severe COVID-19 and COVID-19-related mortality.

The ongoing coronavirus disease 2019 pandemic is spreading relentlessly, and has affected more than 230 million people worldwide. COVID-19 is associated with worse outcomes in the elderly population, and those with comorbid health conditions such as obesity, diabetes mellitus, hypertension, and cardiovascular disorders [1] [2] [3] [4] [5] [6] [7] [8] [9] .

Tuberculosis is a destructive pulmonary disease and therefore widely perceived to be associated with increased susceptibility to acquiring COVID-19, and poorer prognosis in patients having both diseases concurrently, especially among people living with human immunodeficiency virus infection (PLHIV). A note from World Health organization (WHO) also anticipated poorer outcomes in these patients [10] . However, the actual impact of tuberculosis on occurrence and clinical outcomes of COVID-19 is not clear. A case series from Italy reported a benign clinical course for patients having both infections [11] . An early meta-analysis of six Chinese studies found no association between tuberculosis and COVID-19 severity or mortality [12] . Population based data from South Korea also did not suggest tuberculosis to be significantly associated with COVID-19-related mortality [13] . However, other investigators describe a disproportionately higher rate of adverse clinical outcomes among patients with tuberculosis and COVID-19 [14] [15] [16] . Tuberculosis was identified as the commonest comorbidity on verbal autopsy among 70 COVID-19 deaths, and in 10% of whole-body autopsies, in Zambia [17, 18] . Two meta-analyses suggest higher odds of underlying tuberculosis among patients with severe COVID-19 and those dying from COVID-19 [19, 20] . Due to these inconsistencies, we felt a need to perform a detailed analysis of the available evidence till date. Herein, we evaluate the frequency of concurrent active pulmonary tuberculosis among COVID-19 patients. We also assess if comorbid pulmonary tuberculosis increases the risk of severe disease, hospitalization, or mortality in COVID-19 patients.

We registered our systematic review protocol with the PROSPERO database (registration number CRD42021245835). We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and the Meta-analysis of Observational Studies in Epidemiology (MOOSE) recommendations for reporting our review [21, 22] . An approval from our Institutional Review Board was not necessary as we extracted only summary information from previously published articles.

describing original observations, reviews, guidelines, conference abstracts, editorials, and study protocols. Full texts of citations considered potentially suitable by either reviewer were assessed further.

We included a publication for data synthesis if it (a) included patients with COVID-19 confirmed by detection of novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA in respiratory specimens, or strongly suspected on clinical or radiological assessment if a confirmatory test was not available, (b) either described the frequency of patients having concurrent active pulmonary tuberculosis among COVID-19 patients, or reported on any of the following outcomes in COVID-19 patients with and without tuberculosis-severe COVID-19, hospital admission, or mortality. Severe COVID-19 was defined based on institutional or national guidelines, or as per the prevalent guidance from international professional bodies or the World Health Organization. If the same (or substantially overlapping) patient cohort was reported in two or more publications, we included the one describing the largest patient population. In case of any disagreement, consensus between the two reviewers determined study inclusion.

We obtained information on study design, location and healthcare setting, participant inclusion and exclusion criteria, period of patient enrollment, the source of patient information, and the outcomes reported, from all eligible studies. We used the Newcastle-Ottawa Scale (NOS) to assess methodological quality of studies [24] . We considered a study to be of good quality if its NOS score was seven or more (out of a maximum possible score of nine).

We estimated the percentage of active tuberculosis patients among those with COVID-19 disease in each study and calculated the corresponding 95% confidence interval (95% CI) by Clopper-Pearson exact method [25] . We also computed the relative risk (RR), and the corresponding 95% CI, for each predefined outcome from each study [26] . We employed a continuity correction of 0.5 for studies having 'zero' cell frequencies prior to these calculations.

We pooled our data using the DerSimonian-Laird random effects model to generate summary estimates [27] . Freeman-Tukey double arcsine transformation was used to summarize data on proportions [28] . We assessed between-study heterogeneity through the Higgins' inconsistency index (I 2 ), which was considered high for values greater than 0.75 [29] . The contribution of each study to overall heterogeneity, and its influence on the summary estimate, were assessed through the Baujat's plot [30] . For searching the reasons for heterogeneity, we performed subgroup analyses for predefined covariates that included study location and setting, study design, COVID-19 diagnostic standards, description of criteria used to define active tuberculosis, national burden of tuberculosis, and the overall study quality. World Health Organization standards were used to refer to countries as high burden, and to extract country incidence estimates, for tuberculosis [31] . In a sensitivity analysis, the influence of each study on the summary estimate was also assessed by repeating meta-analysis after iteratively omitting one study at a time. Further, any influential study was identified using a battery of diagnostic tests using Studentized Residuals, Difference in Fits (DFFITS), Cooks Distance, Covariance Ratio, Tau Square, and the contribution of each study in the Q, H2 test statistics value and the weights assigned to these studies [32] . Publication bias was assessed through Eggers' test and by visualizing contour-enhanced trim-and-fill funnel plots [33, 34] . We utilized the statistical softwares Stata (Intercooled edition 12.0, Stata Corp, USA) and R (version 4.1.1, R Foundation for Statistical Computing, Austria) for analyzing our data.

We identified 3,375 publications from our literature search (Fig 1) . We finally selected 43 studies, describing 236,863 patients with COVID-19, for data synthesis . Thirty-three (76.7%) of them provided information on one or more of the adverse clinical outcomes of interest (Table 1 ). There were 30 (69.8%) publications from Asia, and 10 (23.3%) from Africa, with maximum contribution from China (22 studies) ( Table 1) . One (2.3%) study analyzed data from multiple countries [43] . All studies evaluated data from retrospective patient cohorts, except for four (9.3%) that collected the information prospectively [39, 40, 56, 69] . Six (14.0%) studies reported population-based data [35, 36, 42, 52, 59, 62] , while the others were conducted in a hospital setting. Two (4.7%) studies also included COVID-19 patients based on high clinical or radiological suspicion [39, 44] . All others only studied patients with disease confirmed by the detection of SARS-CoV2 RNA in respiratory specimens. One (2.3%) study did not specify the inclusion criteria [46] . Only two (4.7%) studies specifically evaluated children [63, 71] ; others included only adults or described a mixed population. Patient information was retrieved mainly from medical records at participating healthcare facilities, or from surveillance registries or insurance databases (Table 1) . Most investigators reviewed patient records or used tuberculosis-related diagnostic codes in databases to identify patients having active tuberculosis (Table 1) . Fourteen (32.6%) studies reported human immunodeficiency virus (HIV) seroprevalence in their patient cohorts [36, 37, 42, 48-50, 58-60, 62, 64, 67, 69, 71] . Of these, a single study from South Africa provided tuberculosis prevalence and outcome data based on HIV status [36] . Only six (14.0%) studies were considered high quality (S1 Table) 

The proportion of patients having active pulmonary tuberculosis among COVID-19 patients could be computed from all 43 studies. It ranged from 0.18% to 14.42% (Fig 2) . The highest occurrence was noted in a study conducted in a high HIV prevalence South African setting [60] . All other studies described figures below 6%. Almost all studies reported proportion estimates of comorbid pulmonary tuberculosis among COVID-19 patients that were higher than their corresponding WHO country estimates for annual tuberculosis incidence (Fig 2) . The pooled proportion estimate from all 43 studies was 1.07% (95% CI 0.81%-1.36%). There was substantial heterogeneity between the studies (I 2 94.7%). Baujat's plot suggested that three studies unduly influenced heterogeneity as well as pooled estimates (S1 Fig) [36, 59, 62] . Omitting these three studies from meta-analysis resulted in a slightly higher summary estimate of proportion (1.15%, 95% CI 0.84%-1.50%) with only a minor reduction in heterogeneity (I 2 86.9%). On influence analysis, a single study with the highest reported proportion of active tuberculosis patients was associated with large values of Studentized residuals, Cook's distance and DFFITS (S2 Fig), and was considered potentially influential [60] . After removing this study, the pooled proportion estimate from remaining 42 studies was lower at 1.00% (95% CI 0.75%-1.28%) with hardly any reduction in heterogeneity (I 2 94.5%). On sensitivity analysis, omitting other studies one at a time also did not appreciably influence summary estimates or heterogeneity (S3 Fig). On subgroup analysis, studies conducted in low tuberculosis burden or multiple countries showed lesser heterogeneity (Table 2) . Overall, the pooled estimates on proportion were much lower from studies conducted in countries not having high tuberculosis burden, as well as from population-based and high-quality studies ( Table 2) .

Twenty studies with 24,371 COVID-19 patients, of whom 161 (0.7%) had tuberculosis, provided information on severe COVID-19 [35, 38, 42, 44, 46, 47, 49, 51, 53-58, 70, 73-77] . All, except four (20.0%), of these publications were from China [35, 42, 49, 58] . Severe COVID-19 was defined based on World Health Organization guidance in three (15.0%) studies [35, 49, 58] , recommendations from international professional bodies in two (10.0%) studies [44, 70] , national guidelines in 11 (55.0%) studies [38, 42, 46, 47, 51, 53, 56, 57, 73, 74, 76] , and institutional policy in four (20.0%) studies [54, 55, 75, 77] . Only three (15.0%) studies were considered high quality [42, 54, 58] . All studies, except two, included patients with laboratory confirmed COVID-19 [44, 46] . Only one (12.5%) had a prospective study design [56] . Of the 3431 patients with severe disease in the included cohorts, 36 (1.0%) had underlying tuberculosis. Only four (20.0%) studies reported a RR for severe COVID-19 that significantly exceeded 1.0 (Fig 3) [42, 46, 56, 77] . COVID-19 patients who also had tuberculosis were 1.46 (95% CI 1.05-2.02) times more likely to develop severe COVID-19 as compared to COVID-19 patients without tuberculosis (Fig 3) .

There was moderate heterogeneity between the studies (I 2 42.9%). Baujat's plot indicated that one Korean study unduly influenced heterogeneity as well as pooled estimates [42] . No additional influential study was identified on formal influence analysis. Omitting this single study from analysis resulted in a lower summary RR estimate (1.35, 95% CI 0.98-1.87) and lesser heterogeneity (I 2 29.5%). On sensitivity analysis, omitting other studies one at a time did not significantly affect heterogeneity (S4 Fig). On subgroup analysis, studies conducted in Africa or in low tuberculosis burden or multiple countries, as well as population-based studies, showed negligible heterogeneity (Table 3) . Overall, the pooled RR estimates were much higher from studies conducted in countries not having high tuberculosis burden, as well as from population-based studies (Table 3 ). There was no significant publication bias (S5 Fig) . 

Four publications with 28,438 COVID-19 patients, of whom 479 (1.7%) had tuberculosis, provided data on hospitalization due to COVID-19 [36, 45, 52, 71] . All these studies were from high tuberculosis burden countries (two from South Africa, and one each from China and Philippines), had a retrospective study design, and included patients with laboratory confirmed COVID-19. Two (50.0%) studies were of high quality [36, 52] . Overall, 20.6% of patients were hospitalized. Of the 5853 patients who required hospitalization in the included cohorts, 227 (3.9%) had underlying tuberculosis. Two studies reported a RR for hospitalization that statistically significantly exceeded 1.0 (Fig 3) [36, 71] . COVID-19 patients who also had tuberculosis were 1.86 (95% CI 0.91-3.81) times more likely require hospitalization as compared to COVID-19 patients without tuberculosis (Fig 3) . This pointed to the absence of any statistically significant risk of hospitalization among COVID-19 patients with tuberculosis.

There was considerable heterogeneity between the studies (I 2 97.5%). A subgroup analysis was not undertaken due to small number of studies. There was no significant publication bias (S5 Fig) .

Seventeen studies with 42,321 COVID-19 patients, of whom 632 (1.5%) had tuberculosis, reported on deaths due to COVID-19 [36, 39, 41-43, 45, 47-50, 52, 57, 60, 63, 65-67] . All studies were conducted in high tuberculosis burden countries, except one from South Korea and another that combined data from multiple nations [42, 43] . All studies, except one, included patients with laboratory confirmed COVID-19 [39] . Only one publication had a prospective study design [39] . Four (23.5%) studies were considered as high quality [36, 42, 43, 52] . Of the 2822 patients who died in the included cohorts, 97 (3.4%) had underlying tuberculosis. Only four (23.5%) studies reported RR for mortality that clearly exceeded 1.0 (Fig 3) [36, 42, 45, 52] . The confidence limits for all other studies were wide (Fig 3) . COVID-19 patients who also had tuberculosis were 1.93 (95% CI 1.56-2.39) times more likely to die as compared to COVID-19 patients without tuberculosis (Fig 3) . There was only mild heterogeneity between the studies (I 2 21.5%). Inspection of Baujat's plot indicated that four studies unduly affected heterogeneity as well as pooled estimates [36, 42, 52, 67] . Two of these were also considered to be influential on formal influence analysis [52, 67] . Omitting these four studies from analysis lowered the summary RR estimate (1.65%, 95% CI 1.25-2.18) and resulted in negligible heterogeneity (I 2 0.0%). On sensitivity analysis, heterogeneity could also be further decreased by individually omitting three of these studies one at a time ( S4 Fig) [36, 42, 67] . Stratification by whether criteria for tuberculosis definition were specified in the studies resulted in homogeneity in either group (Table 3) . On subgroup analysis, studies conducted outside Asia or Africa or in low tuberculosis burden or multiple countries, population-based studies, those including patients with a clinico-radiological diagnosis of COVID-19, and low-quality publications showed negligible heterogeneity (Table 3) . There was no significant publication bias (S5 Fig).

We found that 0.99% of the COVID-19 patients had active pulmonary tuberculosis. These patients showed higher risk for mortality, but not for severe disease or hospitalization, than COVID-19 patients without tuberculosis. Our data synthesis summarizes far greater number of studies than previous meta-analyses and provides information both on tuberculosis frequency and COVID-19 outcome estimates [12, 19, 20] . Unlike previous meta-analyses that reported summary odds ratios, we present summary RR estimates for adverse clinical outcomes, which are much easier to interpret and understand in a clinical setting.

The summary proportion of those with active pulmonary tuberculosis among COVID-19 patients appears higher than the recent WHO estimates for annual incidence of tuberculosis in some of the high tuberculosis burden countries where most of the studies were conducted (China 0.06%, India 0.19%, Nigeria 0.22%, Philippines 0.55%, and South Africa 0.61%) [31] . However, this proportion of active tuberculosis is lower than the generally reported proportion of other comorbid conditions, like diabetes or hypertension [5, 8] . Whether the lower [1] [2] [3] . It is likely that superadded COVID-19 pneumonia in a lung that is already structurally damaged by tuberculosis may manifest as more severe disease. Importantly, local alterations in lung immunity resulting from active pulmonary tuberculosis can also adversely influence host response to SARS-CoV-2 virus. Recent in-vitro data from COVID-19 patients with active pulmonary tuberculosis has shown an attenuated interferon-gamma response after stimulation of whole blood with peptides derived from SARS-CoV-2 spike protein, in contrast to a normal response to Mycobacterium tuberculosis-specific antigens [78] . There are several similarities between COVID-19 and pulmonary tuberculosis. In several countries, COVID-19 too is a stigmatizing disorder, much like tuberculosis. Both diseases show airborne transmission when people are in close contact. Both present with similar symptoms like fever and cough. This can complicate decision-making, especially is nations with high tuberculosis burden. Although several countries have proposed bidirectional screening of both COVID-19 and pulmonary tuberculosis patients, such policy remains difficult to implement in resource-constrained settings. This might contribute to underdiagnosis of tuberculosis in COVID-19 patients. As it is, under-reporting of tuberculosis is a problem that is globally recognized. This is further compounded by reduced access to tuberculosis diagnosis and treatment as a result of COVID-19 related restrictions. It is therefore possible that our calculations regarding pulmonary tuberculosis among COVID-19 patients might be an underestimate.

Our systematic review has a few limitations. Due to the dynamic nature of the pandemic, and the lag between data collection and publication of results, most studies provide information from the initial months of 2020 and from regions that were severely afflicted earlier. Thus, the figures may not truly represent the patient data from all the geographic locations. Also, most of the included studies had a retrospective design, and collated data from medical records that were likely completed in an overwhelmed health system. This could have resulted in both underreporting as well as misclassification of comorbid health conditions. Several studies reported only on inpatients who have a higher probability of adverse outcomes compared to patients in the community. Only 15.6% of the included studies were of sufficiently high quality. There were differences in healthcare strategies regarding SARS-CoV-2 testing and admission/transfer criteria, variability in institutional practices in the timing of investigations and other evaluations, and the level and extent of medical intervention available to patients. Such heterogeneity can restrict the generalizability of our results. We cannot rule out an overestimation from lack of adjustment for potential confounders (like age, HIV status, other comorbid health conditions, or other patient characteristics) as we focused on univariate estimates. In particular, only one South African study reported on tuberculosis frequency data and outcome parameters stratified by HIV status, and there is need to gather more information on the impact of HIV on COVID-19 and tuberculosis associations.

In summary, the available evidence suggests that COVID-19 patients show relatively higher proportion of concurrent active pulmonary tuberculosis. Active pulmonary tuberculosis significantly increases the risk of severe COVID-19 and COVID-19-related mortality.

Supporting information S1 (7) 

Association of sex, age, and comorbidities with mortality in COVID-19 patients: a systematic review and meta-analysis

Factors associated with disease severity and mortality among patients with COVID-19: A systematic review and metaanalysis

Prevalence and predictors of death and severe disease in patients hospitalized due to COVID-19: A comprehensive systematic review and meta-analysis of 77 studies and 38,000 patients

Predictors of adverse prognosis in COVID-19: A systematic review and meta-analysis

Prevalence of underlying diseases in died cases of COVID-19: A systematic review and meta-analysis

Predictors of in-hospital COVID-19 mortality: A comprehensive systematic review and meta-analysis exploring differences by age, sex and health conditions

Factors associated with COVID-19-related death using OpenSAFELY

Prevalence of comorbidities and its effects in patients infected with SARS-CoV-2: a systematic review and meta-analysis

Cardiovascular risk factors and mortality in hospitalized patients with COVID-19: systematic review and meta-analysis of 45 studies and 18,300 patients

Information Note: Tuberculosis and COVID-19

Clinical characteristics of COVID-19 and active tuberculosis co-infection in an Italian reference hospital

Association between tuberculosis and COVID-19 severity and mortality: A rapid systematic review and meta-analysis

Impact of coronavirus disease-2019 on chronic respiratory disease in South Korea: an NHIS COVID-19 database cohort study

Tuberculosis, COVID-19 and migrants: Preliminary analysis of deaths occurring in 69 patients from two cohorts

Active tuberculosis, sequelae and COVID-19 co-infection: first cohort of 49 cases

Mortality due to TB-COVID-19 coinfection in India

Covid-19 deaths in Africa: prospective systematic postmortem surveillance study

Post-mortem examination of hospital inpatient COVID-19 deaths in Lusaka, Zambia-a descriptive whole-body autopsy series

Implications of COVID-19 in high burden countries for HIV/TB: A systematic review of evidence

An updated meta-analysis on the association between tuberculosis and COVID-19 severity and mortality

The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Annals of internal medicine

Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group

Compendium of TB/COVID-19 studies

The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses

The use of confidence or fiducial limits illustrated in the case of the binomial

Calculating confidence intervals for relative risks (odds ratios) and standardised ratios and rates

Meta-analysis in clinical trials

Transformations related to the angular and the square root

Measuring inconsistency in meta-analyses

A graphical method for exploring heterogeneity in meta-analyses: application to a meta-analysis of 65 trials

Geneva: World Health Organization

Outlier and influence diagnostics for meta-analysis

Bias in meta-analysis detected by a simple, graphical test

Contour-enhanced meta-analysis funnel plots help distinguish publication bias from other causes of asymmetry

Epidemiological investigation of the first 5685 cases of SARS-CoV-2 infection in Qatar

Risk factors for COVID-19 death in a population cohort study from the Western Cape Province, South Africa

Clinical characteristics and outcomes of older patients with Coronavirus disease 2019 (COVID-19) in Wuhan, China: A single-centered, retrospective study

Temporal changes of CT findings between non-severe and severe cases of COVID-19 pneumonia: a multi-center, retrospective, longitudinal Study

Predictors of mortality for patients with COVID-19 pneumonia caused by SARS-CoV-2: a prospective cohort study

A profile of a retrospective cohort of 22 patients with COVID-19 and active/treated tuberculosis

Epidemiology of COVID-19 and predictors of outcome in Nigeria: A single-center study

Clinical characteristics and risk factors for fatality and severity in patients with Coronavirus disease in Korea: A nationwide population-based retrospective study using the Korean Health Insurance Review and Assessment Service (HIRA) database

Mortality risk of COVID-19 in elderly males with comorbidities: a multi-country study

Risk factors for severity and mortality in adult COVID-19 inpatients in Wuhan

Clinical outcomes of COVID-19 in Wuhan, China: a large cohort study

Correlation of laboratory index and clinical classification among 342 corona virus disease 2019 patients in Ezhou

Association of overlapped and un-overlapped comorbidities with COVID-19 severity and treatment outcomes: a retrospective cohort study from nine provinces in China

Factors associated with COVID-19 hospital deaths in Espirito Santo, Brazil, 2020

Clinical characteristics and outcomes of patients hospitalized for COVID-19 in Africa: Early insights from the Democratic Republic of the Congo

High HIV prevalence in an early cohort of hospital admissions with COVID-19 in Cape Town

Characteristics and prognostic factors of disease severity in patients with COVID-19: The Beijing experience

Previous and active tuberculosis increases risk of death and prolongs recovery in patients with COVID-19

The clinical features of the 143 patients with COVID-19 in the northeast region of Chongqing

D-dimer level is associated with the severity of COVID-19

South Africa

Communicable and noncommunicable co-morbidities and the presentation of COVID-19 in an African setting of high HIV-1 and tuberculosis prevalence

Clinical presentations, pregnancy complications, and maternal outcomes in pregnant women with COVID-19 and tuberculosis: A retrospective cohort study

COVID-19-related laboratory analyte changes and the relationship between SARS-CoV-2 and HIV, TB, and HbA1c in South Africa

Impact of comorbidities on outcome in children with COVID-19 at a tertiary care pediatric hospital

The Febrile Emergency Unit at Muniz Hospital facing COVID-19, HIV and tuberculosis

Development and external evaluation of predictions models for mortality of COVID-19 patients using machine learning method

Clinical characteristics and predictors of mortality in young adults with severe COVID-19: a retrospective observational study

Risk factors for secondary hemophagocytic lymphohistiocytosis in severe coronavirus disease 2019 adult patients

High prevalence of diabetes and other comorbidities in hospitalized patients with COVID-19 in Delhi, India, and their association with outcomes

Bacterial infection, antibiotic use and COVID-19: Lessons from the intensive care unit

Distinct effects of asthma and COPD comorbidity on disease expression and outcome in patients with COVID-19

Clinical experience with severe acute respiratory syndrome Coronavirus 2-related illness in children: hospital experience in Cape Town, South Africa

Clinical characteristics and outcome of critically ill SARS-CoV-2 patients admitted to intensive care unit of a tertiary care hospital in north india: A retrospective study

Relationship between blood eosinophil levels and COVID-19 mortality

Low serum calcium and phosphorus and their clinical performance in detecting COVID-19 patients

Predictors of clinical deterioration in nonsevere patients with COVID-19: a retrospective cohort study

Analysis of COVID-19 epidemic and clinical risk factors of patients under epidemiological Markov model

An interpretable model-based prediction of severity and crucial factors in patients with COVID-19

Coinfection of tuberculosis and COVID-19 limits the ability to in vitro respond to SARS-CoV-2