key: cord-0818549-5fol0fqb authors: Ulloque-Badaracco, Juan R.; Hernandez-Bustamante, Enrique A.; Herrera-Añazco, Percy; Benites-Zapata, Vicente A. title: Prognostic value of apolipoproteins in COVID-19 patients: A systematic review and meta-analysis date: 2021-11-06 journal: Travel Med Infect Dis DOI: 10.1016/j.tmaid.2021.102200 sha: 008022cf9e8a00b779d8e8816af24e722b96fc98 doc_id: 818549 cord_uid: 5fol0fqb INTRODUCTION: Apolipoproteins are predictive biomarkers for cardiovascular, neoplasms and cerebrovascular diseases and are postulated as prognostic biomarkers in infectious diseases, as COVID-19. Thus, we assessed the prognosis value of apolipoproteins for COVID-19 severity and mortality. METHODS: We conducted a systematic review and meta-analysis using observational studies that reported the association between apolipoproteins and severity or mortality in COVID-19 patients. Newcastle-Ottawa was used for the quality assessment of included studies. Effects measurements were shown as odds ratios (ORs) with 95% confidence intervals (CIs), and Egger-test was developed for assessing the risk of bias publication. RESULTS: We analyzed 12 cohort studies (n = 3580). Patients with low ApoliproteinA1 (ApoA1) (OR 0.35; 95%CI 0.24 to 0.49; P < 0.001) and ApoliproteinB (ApoB) (OR = 0.78; 95%CI 0.69 to 0.87; P < 0.001) values had a higher risk of developing severe disease. ApoB/ApoA1 ratio showed no statistically significant association with higher odds of severity. Low ApoA1 levels were associated with higher odds of all-cause mortality (OR = 0.34; 95%CI 0.20 to 0.57; P < 0.001). ApoB values showed no statistically significant association with a high risk of all-cause mortality. CONCLUSION: We suggest that adequate levels of ApoA1 and ApoB can be a protective factor for severity in COVID-19, and ApoB/ApoA1 ratio did not show predictive utility for severity. Since new coronavirus disease 2019 (COVID-19) was declared as pandemic and a global health emergency by World Health Organization (WHO) [1] , clinical research has been focused on describing the natural history of disease and setting effective treatments, as well as on developing vaccines for COVID-19. As a result, it has provided evidence for diagnosis criteria, categorizing patients with a higher risk of poor outcomes, and suitable allocation of resources, especially for middle-income and low-income countries. This way, clinical research has been able for health systems to manage COVID-19 patients optimally and build an evidence-based treatment. Nonetheless, the emergence of new SARS-CoV-2 variants with higher associated mortality and the likelihood that SARS-CoV-2 will become an endemic virus constitutes an uncertain future for the population and mainly for health staff [2, 3] . Furthermore, daily COVID-19 patient care requires routine laboratory examinations and specific laboratory profiles for underlying diseases. As a result, laboratory tests can be helpful to classify patients according to their risk of progression, prognosis, treatment strategies and other objectives [4] . In this sense, several biological markers and proportions derived from them have been evaluated as indicators of severity and mortality in COVID-19 patients, such as D-dimer [5] [6] [7] , C-reactive protein [8, 9] , neutrophil to lymphocyte ratio [10] , apolipoproteins [11] , albumin to globulin ratio [12] , among others. In the case of ApoA1, ApoB and the ApoA1/ApoB ratio, its prognostic value in cardiovascular and cerebrovascular diseases and neoplasms is known. Although there are studies that suggest its prognostic value in sepsis and bacterial diseases, to our knowledge, the evidence of its prognostic value in patients hospitalized for COVID-19 was not systematized. Recently, these biological markers have been evaluated as prognosis indicators of severity in COVID-19 patients [13] [14] [15] . In order to keep increasing knowledge about COVID-19 and supporting clinical practise, we conducted a meta-analysis of available evidence to assess the prognosis value of apolipoproteins and ApoB/ApoA1 ratio for COVID-19 severity and mortality. This systematic review was registered on the International Prospective Register of Systematic Reviews (PROSPERO) with code CRD42021274326, and the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) statement was followed for reporting [16] . The research question was based on Population, Exposure, Comparison and Outcome (PECO) strategy: Do COVID-19 patients (P) with low values of apolipoproteins (E) have more risk of severity or all-cause mortality (O) compared to normal values of apolipoproteins (C)? The Peer Review of Electronic Search Strategies (PRESS) [17] checklist was used for building the search strategy, and no language or date restriction was applied. On August 28, 2021, a systematic search was performed for retrieving studies assessing the association between Apolipoprotein A1 (ApoA1), Apolipoprotein B (ApoB) or ApoB/ApoA1 ratio, and severity of COVID-19, through the following peer review databases: Embase, PubMed, Web of Science, Scielo, Scopus, LILACS and The Cochrane Library. In addition, a manual search was carried out in preprint databases (Medrixv, Scielo Preprints and ResearchSquare) and other sources (Wangfang Database and CNKI databases). At first, a search strategy based on MeSH and free terms was built for Pubmed, and it was adapted to the other databases (see Search Strategy in Appendix 1 of the Supplemental Information). We included studies with a case-control or cohort design, conducted in patients aged more than 18 years old with a confirmed COVID-19 diagnosis, and assessed the association between ApoA1, ApoB or ApoB/ApoA1 ratio values reported at hospital admission and COVID-19 severity or mortality. Duplicates and studies without all eligibility criteria were excluded. The primary outcome was the severity of COVID-19, and mortality was a secondary outcome. COVID-19 severity was defined as meeting at least one of the following criteria: shortness of breath, respiration rate (RR) ≥ 30 times per minute, blood oxygen saturation at rest ≤93%, PaO2/FiO2 ≤ 300 mmHg or ICU admission. However, definitions for severity are diverse among studies and could be sources of heterogeneity. Rayyan QCRI software was used for study selection and removing duplicates [18] . First, two authors (JRUB and EAHB) screened the retrieved records independently by titles and abstracts. Then, the same two authors assessed the remaining records independently by full-text. Any conflict in the screening process was resolved by two authors (VABZ, PH-A). Afterwards, two authors collected data from included studies in a preset data extraction Microsoft Excel © sheet (JRUB and EAHB). Collected data were: first author, study title, publication date, study design, study location, population baseline characteristics (number of participants, age, sex, comorbidities, stratified sample data), exposure measurements (mean with standard deviation or median with interquartile range, for ApoA1, ApoB or ApoB/ApoA1 ratio from the overall sample and according to sample stratification) outcome type (severity or mortality) and association measures (crude and adjusted). Quality assessment was evaluated independently with the Newcastle-Ottawa Scale (NOS) [19] by two authors (JRUB and EAHB), and scores greater than or equal to six were categorized as low risk of bias. Publication bias was assessed through funnel plots, Egger's test and the trim-and-fill method [20] . Statistical analysis was performed using Review Manager 5.4 (Rev-Man 5.4) (The Cochrane Collaboration, Copenhagen, Denmark). Continuous data reported as the median and interquartile range (IQR) were transformed into means and standard deviations (SD) according to Wan et al. [21] . In order to analyze continuous values of apolipoproteins, standardized mean differences were converted to the natural logarithm of odds ratio and its standard error following Chinn method [22] . Heterogeneity analysis was assessed using the I 2 test and Cochran's Q-statistic. Test values were categorized as severe heterogeneity (>60%), moderate heterogeneity (40-60%) and mild heterogeneity (<40%). A p-value of <0.05 was considered statistically significant. Due to anticipated heterogeneity, a random-effects meta-analysis was performed. Additionally, a subgroup analysis was carried out by study location (Chinese vs non-Chinese studies), and the interaction test pvalue per subgroup analysis was reported. Finally, sensitivity analyses were performed using the low risk of bias studies only. The comprehensive search strategy identified 432 articles, and 214 studies remained after removing duplicates. The screening process by titles and abstracts left 32 studies for full-text review (see Excluded articles by full-text in Supplemental Table S1 ). In turn, screening by fulltext left 12 studies respecting all eligibility criteria [13] [14] [15] [23] [24] [25] [26] [27] [28] [29] [30] [31] . This process is summarised in a flow chart (Fig. 1) . Collected data from included studies are reported in Table 1 and Table 2 . A total of 12 cohort studies were included, of which nine studies analyzed severity, three studies analyzed mortality, and only one study analyzed both outcomes. In addition, eight studies were conducted in China, two in the United Kingdom, one in France and one in Belgium. The included studies were conducted between 2020 and 2021, with 3580 patients hospitalized for COVID-19, of which only 1305 are male. The age range among the total participants ranged from 33 to 83 years. According to the quality assessment of included studies by NOS, seven studies were at low risk of bias, while the remaining five were at moderate risk of bias (Supplemental Table S2 ). The association was found in 10 studies (n = 3077). We found that COVID-19 patients with low ApoA1 values have a higher risk of developing severe disease (OR 0.35; 95% CI 0.24 to 0.49; P < 0.001) with severe heterogeneity (I 2 = 93%) ( Fig. 2A) . In the analysis of subgroups by study location, differences were found between Chinese studies (OR 0.14; 95% CI 0.06 to 0.36; P < 0.001) and non-Chinese studies (OR 0.71; 95% CI 0.59 to 0.84; P < 0.001) (Fig. 2B ). In the sensitivity analysis for including only articles with a low risk of bias (Fig. 2C) , it was found that the association between ApoA1 and the risk of developing severity is still present (OR 0.81; 95% CI 0.76 to 0.87; P < 0.001), but with null heterogeneity (I 2 = 0%) The association was found in nine studies (n = 1375). We found that COVID-19 patients with low ApoB values have a higher risk of developing severe disease (OR = 0.78; 95% CI 0.69 to 0.87; P < 0.001) with mild heterogeneity (I 2 = 12%) (Fig. 3) . The association was found in five studies (n = 1374). However, no statistically significant association was found when low ApoB/ApoA1 values increased the risk of developing the severe disease due to COVID-19 (OR = 1.18; 95% CI 0.95 to 1.46; p = 0.14) with moderate heterogeneity (I 2 = 49%) (Fig. 4) . The association was found in three studies (n = 554). We found that COVID-19 patients with low ApoA1 values have a higher risk of all-cause mortality (OR = 0.34; 95% CI 0.20 to 0.57; P < 0.001) with no heterogeneity (I 2 = 0%) (Fig. 5) . The association was found in four studies (n = 602). No statistically significant association was found when low ApoB values increase the risk of death of COVID-19 patients (OR = 0.63; 95% CI 0.32 to 1.23; p = 0.17) with moderate heterogeneity (I 2 = 41%) (Fig. 6) . In the association between ApoA1 and ApoB with disease severity, publication bias was found (Egger test <0.1). We corrected the publication bias using the trim-and-fill method (OR = 0.39; 95% CI 0.28 to 0.56 and OR: 0.82; 95% CI: 0.74 to 0.90, respectively) (supplemental Figures S1A and S1⋅B). The present systematic review found evidence to recommend low levels of ApoA1 and Apo B as predictors of disease severity and low levels of ApoA1 as a predictor of mortality in patients hospitalized for COVID-19. Exchangeable Apolipoproteins, including Apo As, Apo E, and ApoCs, are constituents of HDL and triglyceride-rich lipoproteins such as VLDL. The best-studied family members are Apo A-I, the most significant HDL protein, in which an anti-atherogenic effect has been documented [32] . In contrast, non-exchangeable Apolipoproteins, such as Apo B, share a similar sequence and structure and can be reversibly associated with lipid surfaces [33] . Apo A is primarily bound to low-density lipoprotein (LDL) in subjects with average triglyceride values. However, Apo A can also bind to APOB100 or triglyceride particles in dyslipidemic states, called very low and intermediate-density lipoproteins [33] . Due to their potential effects and prominence in different pathologies, apolipoproteins have been studied as predictors of clinical outcomes in some diseases. For example, various systematic reviews associated the Apo E with ischemic and hemorrhagic stroke and a higher risk of worse outcomes in patients with traumatic brain disease [34] [35] [36] [37] [38] . Similarly, Apo C was associated with the risk of ischemic stroke, although a systematic review found no evidence of this association [39] . In the case of Apo A1, probably due to its anti-atherogenic effect, some systematic reviews and meta-analyses sought its association with cardiovascular outcomes. Haji Aghajani M et al., in a review of seventeen case-control studies, found an association between Apo A 1 levels and premature coronary artery disease. However, the authors note the lack of good quality prospective cohort studies [40] . Erqou S et al., in a systematic review of thirty-six studies, found that people with smaller Apo A isoforms have an approximately 2-fold higher risk of coronary heart disease or ischemic stroke than those with larger proteins [41] . As with cardiovascular outcomes, other systematic reviews found evidence of Apo A1 as a diagnostic marker for bladder cancer [42] , a poor prognosis of multiple cancers [43, 44] , and it was found at lower levels in patients with Alzheimer's disease [45] . To the best of our knowledge, no systematic reviews have been published on the association between Apolipoprotein values as a prognostic factor in patients with COVID-19 or some other infectious disease; however, our results are not surprising. Apo-I's presence characterizes High-density lipoproteins (HDL), and their ability to transport cholesterol from peripheral tissues back to the liver gives it a cardioprotective function [46] . Similarly, it has antioxidant, anti-apoptotic, anti-thrombotic, anti-inflammatory or anti-infectious functions and decreases rapidly in patients with sepsis, which could explain our findings [46] . A study in pediatric patients in intensive care for sepsis found that Apo A5 serum levels were significantly lower in patients who died than survivors. Similarly, Apo A5 serum levels were significantly correlated with multiple organ failure, shock, acute kidney injury, acute liver injury, and gastrointestinal dysfunction, although not respiratory failure [47] . In adults, an association was also found between low levels of Apolipoproteins and a poor prognosis in patients with sepsis. Although the mechanisms are not well understood, it is suggested that the association is explained due to increased platelet activation and monocyte activation [48, 49] . In addition, the low levels of Apo A are related to high levels of inflammation [50] , and this being a prognostic marker in patients infected by COVID-19 [51] , its role in the binding and neutralization of lipopolysaccharides in bacterial infections is known [52] . In patients with virus infections, changes in plasma HDL-C levels were reported during infections, where the viruses would take advantage of the HDL lipid transfer activity in host cells [53] . Although the best evidence is in patients with hepatitis C virus and acquired immunodeficiency virus, in the case of patients with COVID-19 infection, a similar theory is suggested [54] . Therefore, the HDL lipid transfer activity mechanism could explain our results as the relationship between viral load and worse prognosis in patients with COVID-19 is known [55] . Similarly, the hypothesis of the relationship between lipoproteins and inflammation and thrombosis was raised. In this way, our findings could explain since the association between thrombosis and the prognosis are known [56] . Finally, due to its known association with brain and cardiovascular disease, it is possible that in patients with COVID-19, the prognostic value of ApoA1 is mediated by the occurrence of these diseases. Indeed, complications including myocarditis, acute myocardial infarction, heart failure, arrhythmias and venous thromboembolic events are described in these patients [57, 58] . Similarly, concerning cerebrovascular complications, episodes of stroke, necrotizing hemorrhagic encephalitis, among others, were reported [59, 60] . Our study is the first systematic review to evaluate the prognostic value of Apo A1, ApoB and the ratio of both in patients with an infectious disease. In addition, our study used the NOS to assess the risk of bias of the included articles, which allowed sensitivity analyses when the association between Apo A1 and Apo B with the severity of the disease of patients hospitalized for COVID-19 was analyzed. Our findings allow us to suggest a potential low-cost prognostic marker in patients hospitalized for COVID-19 that will allow health personnel to prioritize or individualize management strategies in patients with low values of these markers. The main limitation is the clinical and methodological heterogeneity in the studies analyzed, which we assumed a priori. However, heterogeneity was addressed and explained mainly by studies with a high risk of bias and, to a lesser extent, by studies done in China. Also, the small number of participants in some studies could be overrepresented in their weights in the meta-analysis. In addition, we found publication bias, which was addressed using the trim and fill method, which did not change the direction of the effect found in the meta-analysis. Likewise, the studies in this meta-analysis do not evaluate the effect that some sociodemographic and clinic variables may have on Apo A1 and Apo B. Indeed, some studies find that Apo AI was significantly higher and Apo B levels were significantly lower among women in general, but, between women, Apo B levels were higher in post-versus premenopausal women [61] . Other studies show that some parameters present a large interindividual variability of response, which is significantly influenced by cofactors, such as weight or BMI, for apo B and apo E [62] . We did not find studies reported in African, American or Oceanic countries, so the generalizability of the results should be taken with caution. Therefore, cohort studies are necessary for various populations to establish Apolipoproteins' generalizability in the severity prognosis of COVID-19. Finally, the heterogeneity of the quantitative assays of plasma Apolipoproteins was not considered, which has been suggested as a problem previously [63, 64] and still represents a challenge since tools are needed for the characterization and accurate quantification of apolipoproteins, including their diverse array of variant forms, are required to understand their salutary and disease-related roles [65] . We conclude that adequate levels of Apolipoproteins are a protective factor for severity in COVID-19. In contrast, only adequate levels of Apo A1 evidenced a protective effect against mortality in patients hospitalized for COVID-19. Furthermore, our findings showed that the ApoA1/ ApoB ratio did not show more excellent predictive utility for severity. Therefore, apolipoproteins could be included in the clinical assessment of hospitalized patients with COVID-19. However, primary studies are necessary to define the optimal cut-off point for Apolipoproteins according to the profile of the hospitalized COVID-19 patient. This research has been funded by the Universidad Peruana de Ciencias Aplicadas through grant IP008-2016. The authors do not have conflicts of interest. htt ps://www.who.int/director-general/speeches/detail/who-director-general-sopening-remarks-at-the-media-briefing-on-covid The coronavirus is here to stay -here's what that means World Health Organization. Tracking SARS-CoV-2 variants Comorbidities, clinical signs and symptoms, laboratory findings, imaging features, treatment strategies, and outcomes in adult and pediatric patients with COVID-19: a systematic review and meta-analysis dimer level is associated with the severity of COVID-19 D-dimer levels and covid-19 severity: systematic Review and Meta-analysis D-dimer concentrations and COVID-19 severity: a systematic review and meta-analysis C-reactive protein: a promising biomarker for poor prognosis in COVID-19 infection Lactate dehydrogenase and C-reactive protein as predictors of respiratory failure in CoVID-19 patients Prognostic value of neutrophil-tolymphocyte ratio in COVID-19 patients: a systematic review and meta-analysis Performance of serum apolipoprotein-A1 as a sentinel of Covid-19 The diagnostic significance of usual biochemical parameters in coronavirus disease 19 (COVID-19): albumin to globulin ratio and CRP to albumin ratio Alteration of lipid profile and value of lipids in the prediction of the length of hospital stay in COVID-19 pneumonia patients Low serum level of apolipoprotein A1 may predict the severity of COVID-19: a retrospective study Association of serum HDL-cholesterol and apolipoprotein A1 levels with risk of severe SARS-CoV-2 infection The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration PRESS peer review of electronic search strategies: 2015 guideline statement Rayyan-a web and mobile app for systematic reviews The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in metaanalyses n trim and fill" method of accounting for publication Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range A simple method for converting an odds ratio to effect size for use in metaanalysis Metabolic disturbances and inflammatory dysfunction predict severity of coronavirus disease 2019 (COVID-19): a retrospective study Low total cholesterol blood level is correlated with pulmonary severity in COVID-19 critical ill patients Lipid metabolism changes in patients with severe COVID-19 Dyslipidemia is related to mortality in critical patients with coronavirus disease 2019: a retrospective study Lipid profile features and their associations with disease severity and mortality in patients with COVID-19 The metabolic fingerprint of COVID-19 severity Metabolic biomarker profiling for identification of susceptibility to severe pneumonia and COVID-19 in the general population Coronavirus disease 2019 induced inflammatory response are associated with changes in lipid profiles Severity-associated markers and assessment model for predicting the severity of COVID-19: a retrospective study in Hangzhou, China Apolipoprotein structure dynamics Apolipoprotein(a): structure and biology Apolipoprotein E genotype, cardiovascular biomarkers and risk of stroke: systematic review and meta-analysis of 14 015 stroke cases and pooled analysis of primary biomarker data from up to 60 883 individuals APOE genotype and MRI markers of cerebrovascular disease: systematic review and meta-analysis Apolipoprotein E gene polymorphisms are risk factors for spontaneous intracerebral hemorrhage: a systematic review and meta-analysis Systematic review on apolipoprotein E: a strong genetic cause of hemorrhagic stroke Apolipoprotein E4 polymorphism and outcomes from traumatic brain injury: a living systematic review and meta-analysis The risk of cardiovascular events with increased apolipoprotein CIII: a systematic review and meta-analysis The association between apolipoprotein A-1 plasma level and premature coronary artery disease: a systematic review and meta-analysis Apolipoprotein(a) isoforms and the risk of vascular disease: systematic review of 40 studies involving 58,000 participants Apolipoprotein A1 as a novel urinary biomarker for diagnosis of bladder cancer: a systematic review and meta-analysis Prognostic significance of pretreatment apolipoprotein A-I as a noninvasive biomarker in cancer survivors: a meta-analysis Association between pretreatment serum apolipoprotein A1 and prognosis of solid tumors in Chinese population: a systematic review and meta-analysis Association between serum concentrations of apolipoprotein A-I (ApoA-I) and alzheimer's disease: systematic review and meta-analysis High-density lipoproteins during sepsis: from bench to bedside Apolipoprotein A-V is a novel diagnostic and prognostic predictor in pediatric patients with sepsis: a prospective pilot study in PICU Changes in HDL-associated apolipoproteins relate to mortality in human sepsis and correlate to monocyte and platelet activation Plasma apolipoprotein CI correlates with increased survival in patients with severe sepsis Human secretory phospholipase A2 mediates decreased plasma levels of HDL cholesterol and ApoA-I in response to inflammation in human ApoA-I transgenic mice Hematological findings and complications of COVID-19 Apolipoprotein A-I: a favorable prognostic marker in infective endocarditis HDL in infectious diseases and sepsis high density lipoproteins: handbook of experimental pharmacology Cholesterol, lipoproteins, and COVID-19: basic concepts and clinical applications SARS-CoV-2 viral load at presentation to hospital is independently associated with the risk of death Thrombosis and coagulopathy in COVID-19 Myocardial injury and COVID-19: possible mechanisms Cardiovascular complications in COVID-19 Neurologic complications of COVID-19. AJEM Stroke in COVID-19: a systematic review and meta-analysis Population frequency distributions of HDL, HDL2, and HDL3 cholesterol and apolipoproteins A-I and B in healthy men and women and associations with age, gender, hormonal status, and sex hormone use: the stanford five city project Effects of apo B and apo E gene polymorphisms on lipid and apolipoprotein concentrations after a test meal International study on the comparability of Apo A-1 and Apo B methods Absolute quantification of plasma apolipoproteins for cardiovascular disease risk prediction Quantitative assays of plasma apolipoproteins None. Supplementary data to this article can be found online at https://doi. org/10.1016/j.tmaid.2021.102200.