Impact of obesity, fasting plasma glucose level, blood pressure, and renal function on the severity of COVID-19: A matter of sexual dimorphism?

AIMS: This study aimed to assess whether body mass index (BMI), fasting plasma glucose (FPG) levels, blood pressure (BP), and kidney function were associated with the risk of severe disease or death in patients with COVID-19.
METHODS: Data on candidate risk factors were extracted from patients' last checkup records. Propensity score-matched cohorts were constructed, and logistic regression models were used to adjust for age, sex, and comorbidities. The primary outcome was death or severe COVID-19, defined as requiring supplementary oxygen or higher ventilatory support.
RESULTS: Among 7,649 patients with confirmed COVID-19, 2,231 (29.2%) received checkups and severe COVID-19 occurred in 307 patients (13.8%). A BMI of 25.0-29.9 was associated with the outcome among women (aOR, 2.29; 95% CI, 1.41-3.73) and patients aged 50-69 years (aOR, 1.64; 95% CI, 1.06-2.54). An FPG ≥ 126 mg/dL was associated with poor outcomes in women (aOR, 2.06; 95% CI, 1.13-3.77) but not in men. Similarly, estimated glomerular filtration rate (eGFR) < 60 ml/min/1.73 m2 was a risk factor in women (aOR, 3.46; 95% CI, 1.71-7.01) and patients aged < 70 years.
CONCLUSIONS: The effects of BMI, FPG, and eGFR on outcomes associated with COVID-19 were prominent in women but not in men.

Introduction

Coronavirus disease 2019 (COVID-19) caused by a novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has spread worldwide and resulted in more than 3.5 million confirmed cases and 1 million deaths as of early October 2020. Most patients with COVID-19 develop mild and self-limiting disease; however, a considerable proportion of patients suffer from severe disease characterized by respiratory or multiorgan failure, or hyperinflammatory syndromes [1]. Multiple factors have been associated with the risk of severe or fatal COVID-19, including hypertension, diabetes mellitus, and chronic cardiopulmonary or renal diseases [2], [3]. Furthermore, obesity has been reported to increase the risk of poor outcome [3], [4]. Although these conditions are frequently associated with disease severity and fatality in other infectious diseases, their degree of association with severe COVID-19 remains unclear.

The impact of baseline health condition on the disease course is difficult to evaluate using medical records, as the information on premorbid status recorded in admission records and epidemiological reports are often inaccurate or incomplete unless they are collected prospectively on purpose. In addition, plasma glucose levels, body weight, and blood pressure (BP) at admission might already deviate from true baseline due to the effects of acute infection with SARS-CoV-2. As a result, it is plausible that baseline health assessment performed prior to the onset of COVID-19 would provide a more accurate representation of risk factors associated with disease severity.

To examine the association between baseline health status and the risk of severe disease in patients with COVID-19, we performed a case-control study, using data from the nationwide registry of COVID-19 cases and from the biennial health checkup database in South Korea.

Materials and methods

This retrospective study compared BP, and metabolic and kidney function parameters between COVID-19 patients with severe disease (including fatalities) and those with mild-to-moderate disease. The study protocol was approved by the institutional review board of the Gil Medical Center, Gachon University College of Medicine and Science (GFIRB2020-118) with a waiver of consent.

Data sources

Health checkup data were extracted from the National Health Insurance Service (NHIS) database, which was linked to the Korea Centers for Disease Control and Prevention (KCDC) COVID-19 patient registry, dedicated to collecting information on all confirmed cases in Korea. All Korean residents are covered by NHIS, which provides universal access to healthcare. In addition, the NHIS provides a biennial health checkup to all beneficiaries aged ≥ 20 years; this database was interrogated for data on baseline health status. The participation rate of NHIS-provided checkups among the eligible population was 78.5% in 2017 [5]. The Korean health checkup data have been previously validated and used in studies that assessed the risk of mortality, cardiovascular events, and diabetes [6], [7], [8]. Information on comorbidities (identified using the International Statistical Classification of Diseases and Related Health Problems, 10th Revision) was extracted from the NHIS reimbursement database. The last date of data entry was May 30, 2020.

Study patients and outcomes

Patients aged ≥ 20 years who were diagnosed with COVID-19 and who received the NHIS health checkup in 2018 or later were included in this study. All included patients were diagnosed with COVID-19 based on nasopharyngeal swab or sputum samples examined with a reverse transcriptase polymerase chain reaction (RT-PCR) for the presence of SARS-CoV-2, as per national guidelines [9].

The primary outcome was death or severe COVID-19, defined as requiring any of the following during hospitalization: supplementary oxygen, high-flow nasal cannula, non-invasive ventilation, mechanical ventilation, or extracorporeal membrane oxygenation.

Premorbid metabolic profiles, blood pressure, and kidney function

Data on body mass index (BMI), fasting plasma glucose (FPG), BP, serum lipid levels, and the estimated glomerular filtration rate (eGFR) were extracted from last checkup records. BP and FPG were categorized using guidelines from the American College of Cardiologists/American Heart Association and the American Diabetes Association, respectively (Supplementary Table 1 ) [10], [11]. BMI was categorized according to the Korean Society for the Study of Obesity guidelines [12]. Total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglyceride (TG) levels were classified with commonly used cutoffs. Finally, eGFR was categorized using the Kidney Disease Improving Global Outcomes guidelines [13].

Table 1

Baseline characteristics of patients with COVID-19, stratified by disease severity.

Characteristic	Total cohort (N = 2231)			Propensity score-matched cohort (N = 1331)
	Severity, No. (%)		P	Severity, No. (%)		P
	Severe (n = 307)	Non-severe (n = 1924)		Severe (n = 293)	Non-severe (n = 1038)
Sex
Male	146 (47.6)	725 (37.7)	0.001	138 (47.1)	445 (42.9)	0.198
Female	161 (52.4)	1199 (62.3)		155 (52.9)	593 (57.1)
Age, years
20–29	3 (1.0)	160 (8.3)	<0.001	3 (1.0)	15 (1.4)	<0.001
30–39	8 (2.6)	225 (11.7)		8 (2.7)	40 (3.9)
40–49	17 (5.5)	365 (19.0)		17 (5.8)	85 (8.2)
50–59	67 (21.8)	610 (31.7)		67 (22.9)	335 (32.3)
60–69	97 (31.6)	400 (20.8)		97 (33.1)	400 (38.5)
70–79	75 (24.4)	137 (7.1)		75 (25.6)	137 (13.2)
≥80	40 (13.0)	27 (1.4)		26 (8.9)	26 (2.5)
Coverage for low income	30 (9.8)	112 (5.8)	0.009	30 (10.2)	82 (7.9)	0.203
Comorbidities
Charlson comorbidity index, mean (range)	4.5 (0–14)	2.5 (0–12)	<0.001	4.6 (0–14)	3.3 (0–12)	<0.001
Diabetes	180 (58.6)	576 (29.9)	<0.001	175 (59.7)	436 (42)	<0.001
Hypertension	184 (59.9)	545 (28.3)	<0.001	171 (58.4)	437 (42.1)	<0.001
Chronic heart disease	137 (44.6)	365 (19.0)	<0.001	131 (44.7)	283 (27.3)	<0.001
Chronic lung disease	211 (68.7)	882 (45.8)	<0.001	203 (69.3)	557 (53.7)	<0.001
Asthma	115 (37.5)	459 (23.9)	<0.001	110 (37.5)	281 (27.1)	<0.001
Chronic liver disease	228 (74.3)	978 (50.8)	<0.001	224 (76.5)	691 (66.6)	0.001
Chronic kidney disease	60 (19.5)	125 (6.5)	<0.001	59 (20.1)	94 (9.1)	<0.001
Cancer	47 (15.3)	135 (7.0)	<0.001	47 (16.0)	110 (10.6)	0.011
Rheumatologic disease	3 (1.0)	9 (0.5)	0.222	3 (1.0)	7 (0.7)	0.466
Chronic neurologic disease	120 (39.1)	320 (16.6)	<0.001	114 (38.9)	259 (25.0)	<0.001
Mortality	42 (13.7)	0 (0.0)	<0.001	37 (12.6)	0 (0.0)	<0.001

Statistical analysis

Study variables were compared between patients with severe and non-severe COVID-19. Baseline demographic and clinical characteristics were compared between patient groups using the χ2 test for categorical variables and Student’s t-test for continuous variables. The Charlson Comorbidity Index (CCI) was calculated using standard methods and compared using the Wilcoxon rank-sum test, as its distribution was non-normal [14]. Adjusted logistic regression models were constructed, based on the following covariates: age, sex, NHIS expanded coverage for low household income, and comorbidities. Comorbidities were categorized into disease groups (Supplementary Table 2), identified from diagnostic codes included at least twice in reimbursement records covering the past 3 years and dated before COVID-19 onset.

To mitigate the impact of confounders, we constructed propensity score-matched cohorts. Propensity scores (PS) of severe COVID-19 risk were calculated using logistic regression with the following covariates: age, sex, coverage for low income, and CCI. Each patient with severe COVID-19 was matched with up to five patients with non-severe disease by greedy neighbor nearest matching. The model was further adjusted in logistic regression analysis, using age, sex, and low-income coverage as covariates.

Subgroup analyses planned a priori were performed for sex and age group. All statistical tests were two-tailed and the threshold for significance was set at P-values < 0.05. All analyses were performed using SAS software (version 9.4; SAS Institute Inc., Cary, NC, USA).

Results

Among 7,649 patients with a confirmed diagnosis of COVID-19, a total of 2,231 (29.2%) patients had received an NHIS health checkup in 2018 or later and were included in the analysis. Patients who had received checkups tended to be older and have more comorbidities than those who did not receive checkups (Supplementary Table 3). Patients with severe COVID-19 or death comprised 13.8% (n = 307) of the total cohort, including 42 deaths (1.9%). Patients who were male, older in age, with low household income, higher CCI, or underlying conditions (except rheumatologic disease) were more likely to develop severe or fatal COVID-19 than patients without these characteristics (Table 1). The PS-matched cohort showed smaller differences in matching variables (age, sex, income, and CCI), but these differences remained significant.

High BMI, classified as class 1 obesity (25.0–29.9), was associated across subgroups with severe and fatal COVID-19 (Fig. 1 ). Overall, patients with a BMI 25.0–29.9 were at a significantly increased risk of severe disease (adjusted odds ratio [aOR], 1.73; 95% confidence interval [CI], 1.22–2.45; P = 0.002) compared to those with a BMI < 23.0, while the effect was not significant in other BMI strata. The association between a BMI 25.0–29.9 and the risk of severe or fatal disease was prominent among women (aOR, 2.29; 95% CI, 1.41–3.73; P = 0.001) but not among men (aOR, 1.28; 95% CI, 0.75–2.19; P = 0.363). However, men with morbid obesity (BMI ≥ 30) showed a marginally higher risk of severe disease (aOR, 2.69; 95% CI, 0.92–7.89; P = 0.071), which was not statistically significant; no such trend was observed among women. Among age groups, there was an association between BMI 25.0–29.9 and risk of severe or fatal COVID-19 for patients aged 50–69 years (aOR, 1.64; 95% CI, 1.06–2.54; P = 0.028).

Fig. 1

Adjusted odds ratios for body mass index and composite of severe COVID-19 or death. PS, propensity score. OR, odds ratio. CI, confidence interval.

The association between FPG and severity of COVID-19 demonstrated a differential effect by sex (Fig. 2 ). Women with FPG ≥ 126 mg/dL had a higher risk of severe COVID-19 compared to women with FPG < 126 mg/dL in both total (aOR, 2.28; 95% CI, 1.26–4.13; P = 0.006) and PS-matched (aOR, 2.06; 95% CI, 1.13–3.77; P = 0.019) cohorts; this effect was not observed among men. The effect of FPG levels differed between age groups. FPG ≥ 126 mg/dL was most strongly associated with disease severity among patients aged < 50 years (aOR, 11.46; 95% CI, 1.30–100.76; P = 0.028 in the PS-matched cohort); concurrently, this association was weaker among patients aged 50–69 years (aOR, 1.77; 95% CI, 1.05–3.00; P = 0.032). In contrast, patients aged ≥ 70 years with higher-than-normal FPG levels showed a lower risk of severe disease.

Fig. 2

Adjusted odds ratios for fasting plasma glucose and composite of severe COVID-19 or death. PS, propensity score. OR, odds ratio. CI, confidence interval.

BP was not associated with the risk of severe or fatal COVID-19 in the total and PS-matched cohorts (Fig. 3 ). However, there was a trend toward an association among men (aOR, 1.01 per mmHg of systolic BP; 95% CI, 1.00–1.03; P = 0.056); in contrast, hypertension classes did not show any significant association. Neither hypertension classes nor systolic BP was associated with disease course among women; a similar lack of association was observed across age groups.

Fig. 3

Adjusted odds ratios for blood pressure and composite of severe COVID-19 or death. PS, propensity score. OR, odds ratio. CI, confidence interval.

A total of 2,205 patients had eGFR calculated in their health checkup results; of these patients 298 (13.5%) had severe disease (Supplementary Table 4). An eGFR < 60 ml/min/1.73 m2 was associated with a higher risk of severe COVID 19 (including fatal disease) (aOR, 2.58; 95% CI, 1.52–4.37; P < 0.001; Fig. 4 ). This association was observed among women (aOR, 3.46; 95% CI, 1.71–7.01; P = 0.001), but not among men (aOR, 1.99; 95% CI, 0.88–4.51; P = 0.100). Moreover, eGFR < 60 ml/min/1.73 m2 and COVID-19 severity were significantly associated among patients aged < 70 years but not among those aged ≥ 70 years (aOR, 1.53; 95% CI, 0.48–4.91; P = 0.242). Our findings were consistent in the overall and PS-matched cohorts and in linear regression analyses.

Fig. 4

Adjusted odds ratios for estimated glomerular filtration rate (GFR) and composite of severe COVID-19 or death. PS, propensity score. OR, odds ratio. CI, confidence interval.

Serum lipid profile was not consistently associated with COVID-19 severity (Supplementary Table 5 and Supplementary Fig. 2). In fact, levels of LDL ≥ 70 mg/dL, HDL < 60 mg/dL, and TG ≥ 175 mg/dL did not affect the risk of severe or fatal disease in the present study. However, there was a trend toward increased risk of severe disease associated with low HDL and high TG levels.

Discussion

In the present study based on a nationwide COVID-19 registry combined with an independent regular health checkup data, the effect of FPG levels and eGFR on the risk of severe or fatal COVID-19 varied between sex and age groups. High FPG levels and low eGFR were associated with severe COVID-19 among women and patients aged < 70 years; however, this association was not observed among men or older patients. Previous studies have reported that case fatality and severity rates were higher among men than among women with COVID-19 [15], [16], [17]. Moreover, official sex disaggregated data from greater than 160 countries and territories have shown that the case fatality rate was lower among women than among men in most countries [18].

The gene coding angiotensin converting enzyme 2 (ACE2) is located on the X chromosome, and estrogen is known to increase the level of ACE2 expression [19], [20]. Although ACE2 is a receptor used for cell entry of SARS-CoV-2, downregulation of ACE2 expression is associated with lung injury caused by respiratory viruses, including SARS-CoV [21], [22]. This apparent paradox could be explained by the anti-inflammatory function of ACE2 through its role in the renin-angiotensin system (RAS). Male sex, older age, and SARS-CoV-2 binding all lead to lower levels of ACE2 expression, which may cause acute lung injury and organ damage through exaggerated angiotensin II signalling [23]. As estrogen is involved in the regulation of ACE2 and its expression level is higher in women, it has been suggested that estrogen levels may have a protective role against respiratory failure and mortality in women with COVID-19 [19], [24], [25]. Nevertheless, metabolic abnormalities associated with altered ACE2 expression might counteract these benefits in women with COVID-19 in a manner analogous to that observed in cardiovascular and renal disease, where the risk in women with and without diabetes is lower and higher than in male counterparts, respectively [26], [27], [28], [29]. It remains unknown whether these documented interactions between sex and metabolic disease-related complications is augmented in patients with COVID-19. Further studies are required on the interaction between sex and COVID-19-related organ damage in patients with specific metabolic characteristics.

In the present study, BMI 25–29.9 was associated with severe or fatal COVID-19 across subgroups, except among males and patients aged < 50 years. Obesity has been reported as a significant risk factor for respiratory failure and mortality in patients with COVID-19 [30], [31], [32]. High levels of ACE2 expression by adipocytes, overactive RAS (characterized by higher level of angiotensin II), impaired baseline pulmonary function, endothelial dysfunction, and higher risk of thromboembolism have been proposed as plausible mechanisms linking obesity with the risk of severe or fatal COVID-19 [33]. Our findings are consistent with those from previous reports and meta-analyses. Furthermore, the present findings suggest that a BMI 25–30, categorized as “class 1 obesity” in Korean guidelines but “overweight” in Western guidelines, is a significant risk factor for severe disease among women and patients aged 50–69 years. The impact of ethnic differences in metabolic parameters on disease severity should be considered in future research.

Interestingly, high BP was not associated with the risk of severe disease in any of the present study subgroups. Hypertension has been associated with mortality risk since early reports on COVID-19 cases confirmed this in China; this association was replicated in subsequent studies [34], [35]. Hypertension alters ACE2 expression, increases the risk of cardiovascular and renal failure, and may cause organ damage; thus, the detrimental effect of hypertension on COVID-19 outcomes seems biologically plausible [36]. However, it remains unclear whether hypertension is an independent rather than a confounding factor for severe COVID-19 [37]. Hypertension is common among older adults and it might reflect their general health status. Our findings suggest that further studies are required to elucidate this uncertainty.

Similarly, serum lipid profiles showed no clear association with the risk of severe or fatal COVID-19. While levels of TG ≥ 175 mg/dL were marginally associated with increased risk of severe disease in men, this effect was not statistically significant. Dyslipidemia considered separately from other components of metabolic syndrome did not affect the risk of severe COVID-19 in the present study. A small number of previous studies investigated cholesterol levels in patients with COVID-19, reporting lower cholesterol levels in patients with severe disease than in those with mild disease [36]. However, the study authors measured serum cholesterol levels after the diagnosis of COVID-19; thus, it remains unclear whether low cholesterol levels were risk factors for COVID-19 or a consequence of the disease. In the present study, serum lipid levels measured before the diagnosis of COVID-19 suggest a lack of association.

Our study has several strengths. First, we used a large, nationwide registry of confirmed COVID-19 cases. South Korea has successfully controlled the COVID-19 epidemic because of an aggressive trace-and-isolate strategy, made possible owing to a large testing capacity, experience from the previous MERS-CoV outbreak, and public cooperation. The dataset used in the present study is unlikely to miss a substantial number of the country’s cases, including those of mild disease, allowing the present study to account for patients who did not require hospitalization, a task often difficult if not impossible in other countries. Second, baseline health status data were extracted from a health checkup database, created before the COVID-19 pandemic. As a result, our data were likely an accurate and objective representation of patients’ health status before they acquired the SARS-CoV-2 infection.

However, this study has some limitations. First, less than a half of patients diagnosed with COVID-19 had data available from a health checkup performed in 2018 or later. Patients who were older or had more comorbidities were more likely to have received checkups; thus the baseline characteristics of study participants differed from those of patients excluded from the study due to the lack of checkup data. Nevertheless, we compared variables of interest in a robust design, adjusting for underlying conditions and using a PS-matched analysis to mitigate confounding. Second, there was a time gap between the last checkup and SARS-CoV-2 infection. Some of our study variables are prone to temporal change and the measurements recorded during checkups might not fully reflect patient status immediately before infection. Finally, we could not account for the effect of treatment in our analysis.

In our retrospective study using a nationwide health checkup database, high FPG levels and low eGFR were significantly associated with the risk of severe COVID-19 (including fatal disease) among women and patients aged < 70 years. High BMI was associated with severe illness among women. No consistent increase in risk was observed in association with high BP or dyslipidemia. These findings suggest that the baseline metabolic characteristics exert differential sex- and age-related effects on disease severity among patients diagnosed with COVID-19.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.