Obesity is associated with severe disease and mortality in patients with coronavirus disease 2019 (COVID-19): a meta-analysis

Background The coronavirus disease 2019 (COVID-19) pandemic has led to global research to predict those who are at greatest risk of developing severe disease and mortality. The aim of this meta-analysis was to determine the associations between obesity and the severity of and mortality due to COVID-19. Methods We searched the PubMed, EMBASE, Cochrane Library and Web of Science databases for studies evaluating the associations of obesity with COVID-19. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using random- or fixed-effects models. Meta-regression analyses were conducted to estimate regression coefficients. Results Forty-six studies involving 625,153 patients were included. Compared with nonobese patients, obese patients had a significantly increased risk of infection. (OR 2.73, 95% CI 1.53–4.87; I2 = 96.8%), hospitalization (OR 1.72, 95% CI 1.55–1.92; I2 = 47.4%), clinically severe disease (OR 3.81, 95% CI 1.97–7.35; I2 = 57.4%), mechanical ventilation (OR 1.66, 95% CI 1.42–1.94; I2 = 41.3%), intensive care unit (ICU) admission (OR 2.25, 95% CI 1.55–3.27; I2 = 71.5%), and mortality (OR 1.61, 95% CI 1.29–2.01; I2 = 83.1%). Conclusion Patients with obesity may have a greater risk of infection, hospitalization, clinically severe disease, mechanical ventilation, ICU admission, and mortality due to COVID-19. Therefore, it is important to increase awareness of these associations with obesity in COVID-19 patients. Supplementary Information The online version contains supplementary material available at 10.1186/s12889-021-11546-6.


Background
On December 31, 2019, the World Health Organization (WHO) was made aware of an outbreak involving several cases of atypical pneumonia. These cases were subsequently identified as being caused by a novel virus belonging to the coronavirus (CoV) family, called severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [1]. On January 30, 2020, the WHO declared an international public health emergency due to infections caused by SARS-CoV-2. On February 20, 2020, the WHO officially named the disease caused by SARS-CoV-2 coronavirus disease 2019   [2,3]. COVID-19 has posed a global health threat, causing an ongoing pandemic in many countries and territories, with approximately 6,287,771 confirmed COVID-19 cases and 379,941 deaths [4] as of June 3, 2020. The number of COVID-19 cases has been rising worldwide, and there is increasing global concern about this outbreak [5].
WHO global estimates indicate that 39% of adults are overweight and 13% are obese [6]. Obesity is an increasing worldwide health concern and is regarded as a critical risk factor for various infections, postinfection complications and mortality from severe infections [7]. Obesity has been shown to have deleterious effects on host immunity, which is the primary cause of an increased risk of infection, especially severe infection [7,8]. Obesity has also been shown to affect lung function in multiple ways that are related to mechanical and inflammatory factors, making obese individuals more likely to suffer from respiratory symptoms and progress to respiratory failure [9].
Accumulating evidence suggests that the group of patients who develop severe COVID-19 may have a higher proportion of obesity than the group with non-severe COVID-19; in some reports, the difference was significant [10][11][12][13]. However, a lack of information regarding the global prevalence of obesity in individuals with COVID-19 remains. Investigating the influence of obesity on COVID-19 is of scientific interest. This investigation aimed to review the relationship between obesity and COVID-19. In doing so, we aim to enhance public awareness of the association between obesity and COVID-19. Furthermore, highlighting the possible associations between obesity and COVID-19 could guide those working to control the COVID-19 pandemic.

Literature search
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Individual Participant Data (PRISMA-IPD) statement was followed for the performance and reporting of this meta-analysis [14]. Our metaanalysis focused on the relationships between obesity and the mortality due to and severity of COVID-19.
PubMed, EMBASE, the Cochrane Library and Web of Science were carefully searched from inception to January 2021 for the terms "COVID-19" and "novel coronavirus" combined with the terms "obesity" and "BMI" as index words. Two investigators (ZC and YY) independently reviewed the identified abstracts and selected articles for full review. Disagreements were resolved by a third investigator (JZ). The search strategy is described in a supplementary file (Supplementary File 1).

Inclusion and exclusion criteria
The inclusion criteria were as follows: (1) patients in the studies had confirmed COVID-19; (2) the body mass index (BMI) values were provided; (3) the comorbidities and severity of disease were provided; and (4) the studies were published in English. The exclusion criteria were as follows: (1) case reports, reviews, letters or nonhuman studies; (2) studies written in a language other than English; and (3) studies with insufficient information. Two investigators (ZC and YY) independently selected studies for inclusion, and disagreements were resolved by a third investigator (JZ).

Data extraction
Data extraction was independently conducted by two authors (ZC and YY) using a standardized data collection form that included the author, year, country, patients, BMI values, and outcomes (infection, hospitalization, severe disease, mechanical ventilation, intensive care unit (ICU) admission, and mortality). The characteristics of these studies are shown in Table 1.

Data synthesis and statistical analysis
All analyses and plots were performed and generated using STATA software (version 12.0, STATA Corp, College Station, TX, USA). Forest plots were used to illustrate the association between obesity and COVID-19 in the selected studies. We pooled the data and calculated the odds ratios (ORs) and 95% confidence intervals (CIs) for dichotomous outcomes, including infection, hospitalization, severe disease, mechanical ventilation, ICU admission, and mortality. The results of the included studies were assessed with random-or fixedeffects models. The I 2 statistic was used to assess the magnitude of heterogeneity-25, 50, and 75% represented low, moderate, and high degrees of heterogeneity, respectively. The choice of the appropriate model was based on the results; a fixed-effects model (inverse variance) was used to pool the data if I 2 was < 50%, and a random-effects model (DerSimonian-Laird) was used if I 2 was > 50% [15]. Funnel plots were used to screen for potential publication bias. To determine the robustness of the results, a sensitivity analysis was conducted with sequential elimination of each study from the pool. The threshold of statistical significance was set to 0.05.

Selected studies and baseline characteristics
Overall, 2874 articles of interest were identified in the initial electronic database searches. A total of 1807 duplicate documents were identified. Of these, 285 full-text articles were considered potentially relevant and further assessed for eligibility. After reviewing the titles and abstracts, 239 articles were excluded because they were not written in English, were case series/reports or reviews, did not contain the full text, or had no reported data. The remaining 46 studies were carefully evaluated in detail; these 46 studies met the inclusion criteria and were finally included (Fig. 1). Of the included studies, 18 reported mortality, 10 reported ICU admission, 8 reported the development of severe disease, 7 reported mechanical ventilation, 7 reported infections, and 5 reported  Table 2. Definition of severe COVID-19 used in each study was shown in Table 3. Study design in the included studies were shown in Supplementary Table 1.

Publication bias and sensitivity analysis
We found no potential publication bias in the studies included in the meta-analysis (Fig. 8). The sensitivity analysis suggested that our results are stable and reliable (Fig. 9).

Discussion
We conducted this meta-analysis to determine whether obesity is a predictor of the COVID-19 severity of and mortality. In the present review, we included 46 articles involving 625,153 patients. Obese patients had a significantly increased risk of infection, hospitalization, severe disease mechanical ventilation, ICU admission, and mortality relative to patients of normal weight.  Mechanisms underlying the association of obesity with the severity of and mortality due to COVID-19 The first mechanism underlying the investigated associations involves adipose tissue (AT). Obesity, usually defined as a BMI > 30 kg/m 2 , is characterized by visceral AT expansion and inflammation [58]. Adipocytes secrete a plenty of factors and hormones that affect many organ systems, including the lungs. Underlying mechanisms of obesity on the severity of COVID-19 may involve abnormalities in the production of adipokines by AT, for example, leptin and adiponectin [59,60]. Leptin as a cytokine can have pro-inflammatory functions that influences both innate and adaptive immune responses by stimulating the production (interleukin (IL)-2 and tumour necrosis factor-alpha (TNF-α)) and suppressing the secretion of IL-4 and IL-5 [61]. In contrast, adiponectin is adipokine that exerts anti-inflammatory actions that inhibits (TNF-α, IL-6, and nuclear factor-κB) and induces (IL-10 and IL-1 receptor antagonist) [61]. Leptin concentrations are increased, whereas adiponectin levels are decreased in obesity [62,63]. The imbalance between leptin and adiponectin may result in the development of dysregulated immune response [64]. The second mechanism involves angiotensin-converting enzyme-2 (ACE-2), COVID-19 utilizes the host ACE2 for binding and entry into host cells. The ACE2 expression is highest in AT. The increase of AT in obese patients increases the expression level of ACE2, which may increase their susceptibility to COVID-19 [65].
Third, impaired lung function and higher level of proinflammatory Cytokines may collaborate to promote the development of respiratory viral infections in patients with obesity. Obesity reduces thoracic wall compliance, resulting in a reduction in functional residual capacity and favor the development of atelectasis [9,66].
Finally, obesity results in physiological lung alterations, such as decreased functional residual capacity and hypoxemia [67]. In addition, obstructive sleep apnoea hypopnea syndrome (OSAHS) increases adverse outcomes of COVID-19 [68]. The etiology of OSAHS is complex, and obesity is one of the main causes of the syndrome. OSAHS is related to obesity. About 60-90% of patients with OSAHS are overweight [69], and the incidence rates of OSAHS in the obese patients is near twice that in normal-weight patients [70].
All of the above mechanisms can reasonably explain how obesity increases COVID-19 severity and mortality.

Implications for strategies to treat patients with obesity
Obesity is a clinical predictor of adverse outcomes in COVID-19 patients; therefore, improved intensive care guidelines for patients with elevated BMI are strongly recommended. Individuals with obesity is an important risk factor for COVID-19, including infection, hospitalization, severe disease, mechanical ventilation, ICU admission, and death. Patients with obesity may require special monitoring. Therefore, obesity patients with COVID-19 require special attention. Additionally,   Obesity aggravates adverse outcomes in COVID-19 patients, and the occurrence of COVID-19 also leads to an increase in obesity. The public control of the COVID-19 outbreak is mainly about controlling human contact, which affects people's behavior to a certain extent and contributes to obesity [71]. Isolation susceptibility to incidence of mental illness [72]. Experiencing loneliness can lead to cut down on physical activity [73]. Regular physical activity is important for maintaining body weight. And as economic conditions decline,  people turn to cheaper foods, which tend to be higher in calories [74]. While more and more people are cooking at home, food stored is likely to be processed to extend its shelf life. Processed foods are associated with more fat, carbohydrate and calorie intake, which is more likely to lead to weight gain than a healthy diet [75]. Preventing obesity is important. Losing weight usually involves increasing physical activity and limiting caloric intake. It is said that individuals complete ≥300 min/  week of physical activity for weight maintenance [76].
People implemented a variety of weight loss strategies, including eating less, increasing physical activity, skipping meals, or taking weight-loss pills or diuretics [77]. Among those trying to lose weight, reducing calorie intake is the most common way [78,79].
One study found that use of metformin was significantly associated with a reduction in COVID-19 mortality [80]. Several reasons might explain this finding. First, metformin reduces the binding of the SARS-CoV-2 to the receptor [81]. Second, metformin inhibits the mTOR signaling pathway, thus reducing SARS-CoV-2 infectivity and COVID-19 mortality [80]. Third, metformin can the inflammatory response [82]. Additionally, metformin reduces the risk of adverse outcomes in COVID-19 patients by reducing BMI and body weight [83].
Due to the extensive spread of COVID-19, enforced confinement has influenced the lives of individuals in many ways, including working behaviours, psychological factors, sedentary activities, and other harmful effects on life habits [84]. Because of increased stress and boredom, people tend to overeat, resulting in the consumption of additional energy/calories and an increased craving for food [85]. In this regard, COVID-19 has contributed to the occurrence of obesity.

Theoretical and practical implications
To the best of our knowledge, this is the first metaanalysis to comprehensively assess obesity and COVID-19 outcomes (infection, hospitalization, severe disease, mechanical ventilation, ICU admission, and mortality). Obesity is a risk factor and predictor of serious disease and is a factor in the need for advanced medical care for COVID-19 patients. Basic research is needed to determine the causal relationship between obesity and adverse outcomes of COVID-19. This study has some limitations. First, some indicators, such as the risk of infection, ICU admission, and mortality, had greater degrees of heterogeneity, and subgroup analyses cannot be performed due to the small number of studies on each indicator. However, the trends were consistent across nearly all forest plots. In addition, many of the included articles did not give specific BMI values, and it is not clear how much a specific unit increase in BMI can increase the severity and mortality rate of COVID-19. Third, because this meta-analysis includes data from
Additional file 1. Full electronic search performed in multiple international databases.
Acknowledgments I am grateful to my department leaders for their great encouragement, support and help to this project.
Authors' contributions JZ coordinated the study. ZC, YY, and JZ conceived the study, contributed to the study design, literature search, figure generation, statistical analysis, outcome synthesis and paper drafting and editing. All authors edited and approved the final version of the manuscript.

Funding
This work was supported by grants from the National Natural Science Foundation of China (82070807, 91749118, 81770775, 81730022), the Planned Science and Technology Project of Hunan Province (2017RS3015) and National key research and development program (2019YFA0801903, 2018YFC2000100).

Availability of data and materials
The datasets used and/or analyzed during the current meta-analysis are available from the corresponding author upon reasonable request.

Declarations
Ethics approval and consent to participate Not applicable, as this is a meta-analysis of previously published papers.

Consent for publication
Not applicable.