Blood Hemoglobin Substantially Modulates the Impact of Gender, Morbid Obesity, and Hyperglycemia on COVID-19 Death Risk: A Multicenter Study in Italy and Spain

Abstract

Background: Hyperglycemia and obesity are associated with a worse prognosis in subjects with COVID-19 independently. Their interaction as well as the potential modulating effects of additional confounding factors is poorly known. Therefore, we aimed to identify and evaluate confounding factors affecting the prognostic value of obesity and hyperglycemia in relation to mortality and admission to the intensive care unit (ICU) due to COVID-19.

Methods: Consecutive patients admitted in two Hospitals from Italy (Bologna and Rome) and three from Spain (Barcelona and Girona) as well as subjects from Primary Health Care centers. Mortality from COVID-19 and risk for ICU admission were evaluated using logistic regression analyses and machine learning (ML) algorithms.

Results: As expected, among 3,065 consecutive patients, both obesity and hyperglycemia were independent predictors of ICU admission. A ML variable selection strategy confirmed these results and identified hyperglycemia, blood hemoglobin and serum bilirubin associated with increased mortality risk. In subjects with blood hemoglobin levels above the median, hyperglycemic and morbidly obese subjects had increased mortality risk than normoglycemic individuals or non-obese subjects. However, no differences were observed among individuals with hemoglobin levels below the median. This was particularly evident in men: those with severe hyperglycemia and hemoglobin concentrations above the median had 30 times increased mortality risk compared with men without hyperglycemia. Importantly, the protective effect of female sex was lost in subjects with increased hemoglobin levels.

Conclusions: Blood hemoglobin substantially modulates the influence of hyperglycemia on increased mortality risk in patients with COVID-19. Monitoring hemoglobin concentrations seem of utmost importance in the clinical settings to help clinicians in the identification of patients at increased death risk.

Keywords: COVID-19; epidemiology; hemoglobin; hyperglycemia; machine learning; mortality; obesity.

Figure 1

Flow chart of study design and inclusion criteria.

Figure 2

Incidence odds ratio of (A) admission to the ICU and (B) mortality from COVID-19 by potential prognostic variables. Data are presented as odds ratio (OR) and 95% confidence intervals (CI). The reference groups (OR=1) for each variable are shown as “-”.: The odds ratios shown are adjusted through a logistic regression model which includes all variables listed. Data are also represented as a forest plot.

Figure 3

Mortality and ICU admission prognostic variables identified from machine learning. Boxplots of the normalized permutation importance obtained from the Boruta algorithm for the potential prognostic variables associated to the (A) mortality from COVID-19 in all subjects (n=3065), (B) admission to the ICU in all subjects (n=3065), (C) mortality from COVID-19 in hospital patients (n=1114), and (D) admission to the ICU in hospital patients (n=1114).

Figure 4

Incidence odds ratio of mortality from COVID-19 in men including iron-related parameters as prognostic variables. Data are presented as odds ratio (OR) and 95% confidence intervals (CI). The reference groups (OR=1) for each variable are shown as “-”.: The odds ratios shown are adjusted through a logistic regression model which includes all variables listed. Data are also represented as a forest plot. Hemoglobin and bilirubin were dichotomized based on the median values.

Figure 5

Incidence odds ratio of mortality from COVID-19 for the final models (n=1,114) according to the median hemoglobin concentrations (13.6 g/dL). (A) In individuals with hemoglobin levels below the median, (B) In individuals with hemoglobin levels above the median, (C) Receiver operating characteristic curve for the logistic regression model in individuals with hemoglobin levels below the median, (D) Receiver operating characteristic curve for the logistic regression model in individuals with hemoglobin levels above the median. Data are presented as odds ratio (OR) and 95% confidence intervals (CI). The reference groups (OR=1) for each variable are shown as “-”.: The odds ratios shown are adjusted through a logistic regression model which includes all variables listed. Data are also represented as a forest plot. AUC, area under the curve.