The association between obesity and ICU admission among COVID-19 patients: A meta-analysis of adjusted risk estimates.

Previous meta-analysis studies have investigated the association between obesity and intensive care unit (ICU) admission among patients with coronavirus disease 2019 (COVID-19), however, the pooled effect sizes were synthesized on the basis of primarily un-adjusted effect estimates, which means that the confounding factors were not considered. It was worth noting that class 3 obesity was significantly associated with an increased risk for ICU admission among COVID-19 patients based on age-adjusted analysis (effect size = 1.45, 95% confidence interval (CI): 1.23–1.70), but not while based on multivariable analysis (effect size = 1.15, 95% CI: 0.96–1.36) [1], which suggests that the association between obesity and ICU admission among COVID-19 patients might be confounded by certain confounding factors. Up to now, a considerable number of studies reporting the association between obesity and ICU admission among COVID-19 patients using multivariable analysis adjusting for confounding factors with inconsistent findings have been published. Therefore, we performed a quantitative meta-analysis to estimate the risk factors-adjusted association between obesity and ICU admission among COVID-19 patients on the basis of adjusted effect estimates reported by all eligibly included studies.

This meta-analysis strictly followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines [2,3]. We systematically searched PubMed, EMBASE and Web of Science databases to identify relevant literature which was published between January 1, 2020 and July 15, 2021. The keywords were used as follows: “coronavirus disease 2019” or “COVID-19” or “severe acute respiratory syndrome coronavirus 2” or “SARS-CoV-2” or “2019 novel coronavirus” or “2019-nCoV” and “obese” or “obesity” or “body mass index” or “BMI” and “intensive care unit admission” or “ICU admission”. The exposure group was defined as COVID-19 patients with obesity and the control group was defined as COVID-19 patients without obesity. The outcome of interest was ICU admission. All peer-reviewed articles published in English language regarding the association between obesity and ICU admission among COVID-19 patients using multivariable analysis were included. Duplications, case reports, corrections, reviews, preprints and studies reporting un-adjusted effect size were excluded.

R software (Version 3.6.3) and Stata software (Version 11.2) were used for all statistical analyses. A random-effects meta-analysis model was applied to synthesize the pooled effect size and 95% CI [4,5]. Higgins I2 test was fitted to assess the inter-study heterogeneity [6]. Sensitivity analysis by omitting each study one time was performed to evaluate the robustness of our results [7]. Publication bias was assessed by Begg's rank correlation test [8]. Statistical significance was considered as P ≤ 0.05.

A total of thirty-eight studies with 902,352 COVID-19 patients were enrolled in this meta-analysis. The general characteristics of the included studies are summarized in Supplementary Table S1. Our meta-analysis based on adjusted effect estimates revealed that obesity was significantly associated with an increased risk for ICU admission among COVID-19 patients (pooled effect size = 1.84, 95% CI: 1.61–2.10, Fig. 1A). Consistent results were observed in the subgroup analyses by sample size (pooled effect size = 1.73, 95% CI: 1.47–2.03 for ≥1000 cases and 2.05, 95% CI: 1.65–2.56 for <1000 cases, respectively), age (pooled effect size = 1.77, 95% CI: 1.46–2.14 for ≥60 years old and 1.89, 95% CI: 1.61–2.22 for <60 years old, respectively), male percentage (pooled effect size = 1.89, 95% CI: 1.43–2.51 for ≥60% and 1.84, 95% CI: 1.56–2.17 for <60%, respectively), region (pooled effect size = 2.04, 95% CI: 1.52–2.74 for Asia; 1.39, 95% CI: 1.25–1.55 for North America; 2.18, 95% CI: 1.40–3.39 for South America and 2.38, 95% CI: 1.84–3.07 for Europe, respectively), cohort type (pooled effect size = 1.74, 95% CI: 1.51–2.01 for retrospective studies and 3.01, 95% CI: 2.23–4.05 for prospective studies, respectively) and setting (pooled effect size = 1.97, 95% CI: 1.61–2.40 for hospitalized patients and 1.66, 95% CI: 1.36–2.04 for all patients, respectively). Sensitivity analysis showed that our results were robust and stable (Fig. 1B). Publication bias was not detected by Begg's test (P = 0.421).

Fig. 1

(A) The forest plot demonstrated the significant association between obesity and the increased risk for intensive care unit (ICU) admission among patients with coronavirus disease 2019 (COVID-19) on the basis of thirty-eight eligible studies with a total of 902,352 cases reporting adjusted effect estimates and (B) Leave-one-out sensitivity analysis showed that our results were robust and stable. * indicates combined effects based on subgroups.

A total of thirty-eight studies with 902,352 COVID-19 patients were enrolled in this meta-analysis. The general characteristics of the included studies are summarized in Supplementary Table S1. Our meta-analysis based on adjusted effect estimates revealed that obesity was significantly associated with an increased risk for ICU admission among COVID-19 patients (pooled effect size = 1.84, 95% CI: 1.61–2.10, Fig. 1A). Consistent results were observed in the subgroup analyses by sample size (pooled effect size = 1.73, 95% CI: 1.47–2.03 for ≥1000 cases and 2.05, 95% CI: 1.65–2.56 for <1000 cases, respectively), age (pooled effect size = 1.77, 95% CI: 1.46–2.14 for ≥60 years old and 1.89, 95% CI: 1.61–2.22 for <60 years old, respectively), male percentage (pooled effect size = 1.89, 95% CI: 1.43–2.51 for ≥60% and 1.84, 95% CI: 1.56–2.17 for <60%, respectively), region (pooled effect size = 2.04, 95% CI: 1.52–2.74 for Asia; 1.39, 95% CI: 1.25–1.55 for North America; 2.18, 95% CI: 1.40–3.39 for South America and 2.38, 95% CI: 1.84–3.07 for Europe, respectively), cohort type (pooled effect size = 1.74, 95% CI: 1.51–2.01 for retrospective studies and 3.01, 95% CI: 2.23–4.05 for prospective studies, respectively) and setting (pooled effect size = 1.97, 95% CI: 1.61–2.40 for hospitalized patients and 1.66, 95% CI: 1.36–2.04 for all patients, respectively). Sensitivity analysis showed that our results were robust and stable (Fig. 1B). Publication bias was not detected by Begg's test (P = 0.421).

In conclusion, this current meta-analysis based on adjusted effect estimates indicated that COVID-19 patients with obesity were at high risk for ICU admission. Further well-designed studies with large sample sizes are warranted to verify our findings.

The following are the supplementary data related to this article.

Supplementary Table S1

General information of the eligible studies included in this meta-analysis.

Supplementary data to this article can be found online at https://doi.org/10.1016/j.ajem.2021.08.054.

Author contributions

Ying Wang conducted literature search, screened studies, extracted data, performed statistical analysis, and drafted the original manuscript. Hongjie Hou co-conducted literature search, screened studies, extracted data, performed statistical analysis, and drafted the original manuscript. Jie Xu co-performed statistical analysis, and contributed in finalizing the manuscript. Yadong Wang conceptualized the original idea, drafted the original manuscript, resolved disagreements, and contributed in finalizing the manuscript. Haiyan Yang supervised the project, conceptualized the original idea, critically reviewed and revised all sections of the manuscript, and contributed in finalizing the manuscript.

Conflicts of interest statement

The authors declare that they have no any potential conflict of interest regarding this submitted manuscript.