Abstract
Background & aims
Coronavirus disease 2019 (COVID-19) patients with severe complications have shown comorbidities with cardiovascular-disease, hypertension and type 2 diabetes mellitus; clinical disorders that share the common metabolic alterations of insulin resistance and dyslipidaemia. A high triglyceride to high density lipoprotein cholesterol (Tg/HDL c) ratio has been associated with reduced insulin sensitivity, metabolic syndrome and adverse cardiovascular events. Our aim in this study was to determine the association between different components of the lipid profile and particularly the Tg/HDL c ratio with severe complications like the requirement of invasive mechanical ventilation in COVID-19 patients.
Methods
We collected demographic, clinical and biochemical data to conduct a cohort study in 43 adult patients with confirmed COVID-19 diagnosis by quantitative polymerase chain reaction (qPCR) at baseline and in the subsequent 15 days. Patients were subjected to a very similar treatment scheme with the JAK1/2 inhibitor ruxolitinib. Descriptive statistics, variable association and logistic regression were applied to identify predictors of disease severity among elements and calculations from the lipid profile.
Results
Patients were aged 57 ± 14 years; 55.8% were male from which 75% required hospitalization and 44.2% were female who 58% were hospitalized. The most common comorbidities were type 2 diabetes mellitus (58%) and hypertension (40%). Hospitalized and critical care patients showed lower HDL c blood levels and increased Tg/HDL c ratio than those with outpatient management and mild/asymptomatic COVID-19. Tg/HDL c ratio correlated with variables of disease severity such as lactate dehydrogenase (LDH) levels (r = 0.356; p < 0.05); National Early Warning Score 2 (NEWS 2) (r = 0.495; p < 0.01); quick sequential organ failure assessment (qSOFA) (r = 0.538; p < 0.001); increased need of oxygen support (r = 0.447; p < 0.01) and requirement of mechanical ventilation (r = 0.378; p < 0.05). Tg/HDL c ratio had a negative correlation with partial oxygen saturation/fraction of inspired oxygen (SaO 2/FiO2) ratio (r = −0.332;p < 0.05). Linear regression analysis showed that Tg/HDL c ratio can predict increases in inflammatory factors like LDH (p < 0.01); ferritin (p < 0.01) and D-dimer (p < 0.001). Logistic regression model indicated that ≥7.45 Tg/HDL c ratio predicts requirement of invasive mechanical ventilation (OR 11.815, CI 1.832–76.186, p < 0.01).
Conclusions
The Tg/HDLc ratio can be used as an early biochemical marker of COVID-19 severe prognosis with requirement of invasive mechanical ventilation.
Keywords: Tg/HDLc ratio, COVID-19, HDL c, Mechanical ventilation
1. Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), also known as coronavirus disease 2019 (COVID-19) is caused by a new positive-sense, single-stranded RNA virus which belongs to the β-coronaviruses lineage [1].
Around 80% of the COVID-19 infected patients are asymptomatic or develop very mild symptoms; however, in the remaining 20% a rapid disease progression leads to severe complications as hypoxemia (dyspnoea, central cyanosis and oxygen saturation lower than 92%), with the risk to develop respiratory distress syndrome; shock and multiple organ failure, increasing the mortality rate to more than 50% [2].
Severe COVID-19 has been linked to a proinflammatory cytokine storm defined as a massive release of tumour necrosis factor alpha (TNFα); interleukins (IL) IL-1β, IL-2, IL-6, IL-8; granulocyte colony-stimulating factor (G-CSF); along with reactive oxygen species and chemokines such as C–C motif chemokine ligands (CCL2, CCL3, CCL5) and interferon gamma-induced protein 10 (IP-10) [2,3].
Little is known about the immunological response against SARS-CoV2; nevertheless, clinical trials with patients infected with other coronaviruses, SARS-CoV and MERS-CoV, shown that increases in the serum concentration of proinflammatory cytokines is linked with lung inflammation and extensive lung injury [4]. The cytokine storm in other coronavirus infections is the result of an increase in neutrophil number; monocyte and macrophage hyperactivation, and lymphocyte Th1/Th17 activation in the adaptive immune system, intensifying the inflammatory response [3].
Trained immunity is a memory-like feature of cells from the innate immune system like monocytes which develop epigenetic changes when Toll-like receptors (TLR) are activated by diverse antigens, promoting increased release of proinflammatory cytokines in a second exposure to an antigen [5]. The enhanced response depends on the ligand concentration and the type of receptor activated during the first antigen exposure [5]. Remarkably, several endogenous ligands like saturated fatty acids, oxidized low-density lipoprotein and advanced glycation end products (AGEs) act as damage-associated molecular patterns (DAMPs) and are recognized by Toll-like receptors (TLR2 and TLR4) [6], activating inflammatory responses.
Patients with metabolic syndrome and insulin resistance commonly have higher concentration of these endogenous ligands; therefore, trained immunity could be involved in the increased synthesis of proinflammatory cytokines by the immune system of patients when infected with SARS-CoV-2 [7].
Compared with mild or asymptomatic COVID-19 patients, individuals with severe complications have higher prevalence of comorbidities such as hypertension, cardiovascular disease and type 2 diabetes mellitus [8]. These comorbidities share the common metabolic alterations of insulin resistance and dyslipidaemia; the later has been linked to severe COVID-19 by Choi et al. [9] and Hariyanto et al. [10].
The triglyceride to High Density Lipoprotein-cholesterol (Tg/HDL c) ratio has been associated with reduced insulin sensitivity [11] and adverse cardiovascular events [12] and has even been considered as a better marker of cardiovascular risk than serum concentrations of high- or low-density lipoproteins cholesterol (HDL c or LDL c) alone [12].
Given the lack of studies about a possible association between COVID-19 severity with different components of the lipid profile and particularly with the Tg/HDL c ratio; which represents an easy and affordable estimation of insulin resistance and cardiovascular risk; we here associate this ratio at the time of COVID-19 diagnosis with biochemical values of severity and the development of complications, identifying it as an early marker of negative prognosis.
2. Materials and methods
2.1. Ethical approval
The research ethics review board at Secretaría de salud de la Ciudad de México (SEDESA) approved the protocol as it represents minimum risk and is in accordance with the WMA Declaration of Helsinki. All participants or their relatives provided written informed consent.
2.2. Study design and population
This is an observational cohort study using clinical data of 43 adult patients, 24 (55.8%) male and 19 (44.1%) female admitted from May to October 2020 in our institution with COVID-19 infection, confirmed by quantitative reverse transcriptase-polymerase chain reaction (qRT-PCR) from nasopharyngeal swab specimens in addition to characteristic symptomatology and computerized thoracic tomography imaging.
Patients who had previous dyslipidaemia treatment, uncontrolled type 2 diabetes mellitus or chronic kidney or liver disease were not included.
All patients received a standard clinical laboratory sampling at the baseline (COVID-19 diagnosis). Biochemical parameters were performed on DxC 700 AU chemistry analyser. All laboratory assays had completed the standardization and certification program. Epidemiologic and demographic features as well as the biochemical lipid profile including triglycerides (Tg), total HDL c and LDL c levels were archived. Tg/HDL c ratio was calculated. Other analyses include: hemogram, renal function tests (C cystatin, blood urea nitrogen, creatinine) and severity biochemical parameters (D-dimer, C reactive protein [CRP], erythrocyte sedimentation rate [ESR], ferritin and lactate dehydrogenase [LDH]). Partial oxygen saturation/fraction of inspired oxygen (SaO 2/FiO2) ratio was evaluated, and the early warning scores National Early Warning Score 2 (NEWS 2); quick sequential organ failure assessment (qSOFA) and Comorbidity-Age-Lymphocyte Count-Lactate dehydrogenase score (CALL score). For the correlation assessment of the lipid profile with levels of respiratory support these were classified as 1. No complementary oxygen supply, 2. Conventional nasal cannula, 3. Conventional mask, 4. High-flow mask, 5. High-flow nasal cannula, 6. Non-invasive ventilation and 7. Invasive mechanical ventilation. Subsequent clinical assessment was done within 15 days of hospitalization or outpatient management. The incidence of acute kidney injury defined according to the KDIGO criteria, requirement of invasive mechanical ventilation, vasopressor support, days of hospitalization and mortality were also assessed.
The criteria for severe COVID-19 were a respiratory rate >30/min; O2 saturation <93% and/or SaO 2/FiO2 <160 (severe ARDS Acute Respiratory Distress Syndrome by Kigali criteria) and for critical care COVID-19 were requirement of invasive mechanical ventilation, shock and/or multiple organ failure. In this study both, outpatient management and hospitalized patients had antipyretic drugs and 5 mg/12 h of the JAK1/2 inhibitor ruxolitinib in patients with no-requirement of mechanical ventilation and 10 mg/1 2 h in patients with mechanical ventilation. All hospitalized patients also had treatment with 6 mg/24 h of dexamethasone for 10 days and low-molecular weight heparin. Patients with critical care COVID-19 were treated at the intensive care unit of our institution.
2.3. Statistical analysis
Descriptive tabulation was done using data about age, sex, weight, comorbidities, biochemistry, acute kidney injury, requirement of invasive mechanical ventilation, vasopressor support, days of hospitalization and mortality. The Shapiro–Wilks test was employed to determine normality of data distribution. Statistical analysis of the categorical data was performed using the Chi-square and Fisher exact test. Student's t-test was used to compare continuous values between two groups in which case data were normally distributed and non-parametric Mann-Witney U test used when data were not normally distributed. Kruskal–Wallis test was used to compare continuous values between three groups in which case data were not normally distributed and followed by a Games-Howell post hoc test. Correlation analysis to study biochemical variables and COVID-19 severity was done using Spearman test. Simple lineal regression analyses were used to investigate if any of the lipid profile components and/or the Tg/HDL c ratio are predictive of the biochemical variables of severity and univariate logistic regression analysis were used to investigate if they are predictive of requirement of invasive mechanical ventilation, acute kidney injury, vasopressor support or death. The cut-off values were calculated in accordance with the receiver operating characteristic (ROC) curves and Youden's index. Finally, a multivariate logistic regression analysis adjusted to age, sex and cut-off points calculated for HDL c and the Tg/HDL c ratio was conducted for the requirement of invasive mechanical ventilation with successive steps backward (conditional) method. Data were analysed using IBM SPSS version 20. Statistical differences were considered significant when p-values were <0.05.
3. Results
Outpatient management were done in 14 individuals while 29 developed severe COVID-19 and were hospitalized. The age was not different between outpatient and hospitalized treatment. Demographic features are shown in Table 1 .
Table 1.
Clinical characteristics of patients with COVID-19.
Variable | Global (n = 43) | Outpatient management (n = 14) | Hospitalized (n = 29) | p |
---|---|---|---|---|
Gender (%) | 0.235 | |||
Male∗∗ | 24 (55.8) | 6 (25) | 18 (75) | |
Female∗∗ | 19 (44.1) | 8 (42.1) | 11 (57.8) | |
Age (years)∗ | 57.19 (±13.9) | 59.8 (±17) | 55.8 (±12.2) | 0.508 |
Comorbidities (%) | ||||
Diabetes∗∗ | 25 (58.1) | 7 (28) | 18 (72) | 0.616 |
Hypertension∗∗ | 17 (39.5) | 6 (35.2) | 11 (64.7) | 0.616 |
NEWS 2∗ | 8 (0–16) | 2 (0–8) | 8 (4–16) | <0.001 |
q SOFA∗ | 1 (0–3) | 0 (0–2) | 1 (0–3) | <0.001 |
SaO 2/FiO2∗ | 265 (81–452) | 438 (419–452) | 180 (81–438) | <0.001 |
CALL Score∗ | 8 (4–13) | 9 (4–12) | 8 (4–13) | 0.905 |
Leukocytes (x10e3/μL)∗ | 8.3 (2.7–21.6) | 4.7 (2.7–9.7) | 10.7 (4.5–21.6) | <0.001 |
Lymphocytes (x10e3/μL)∗ | 1.2 (0.5–7.5) | 1.2 (0.5–2.6) | 1.3 (0.8–7.5) | 0.372 |
NLR∗ | 3.9 (0.9–17.1) | 2.3 (1.5–9) | 4.5 (0.9–17.1) | 0.003 |
CRP (mg/dL)∗ | 9.9 (0.2–39.4) | 3.3 (0.2–10.5) | 17.6 (0.4–39.4) | <0.001 |
LDH (IU/L)∗ | 326 (149–1053) | 219 (149–413) | 398 (159–1053) | 0.003 |
ESR (mm 3/hr)∗ | 45 (18–60) | 27 (18–50) | 45 (27–60) | <0.001 |
Ferritin (μg/L)∗ | 601 (41–3629) | 361 (41–700) | 658 (204–3629) | 0.002 |
D-dimer (ng/mL)∗ | 644 (99–4790) | 99 (99–2760) | 882 (99–4790) | 0.001 |
Fibrinogen (mg/dL)∗ | 508 (250–713) | 408.5 (250–533) | 584 (294–713) | 0.026 |
Table 1: Data were expressed as mean ± standard deviation or median (range).
The value of p was derived from ∗U Mann Whitney test; ∗∗Chi-squared test. Statistical significance p values are shown in bold (p < 0.05).
Abbreviations: CALL Score, Comorbidity-Age-Lymphocyte count-Lactate dehydrogenase score; CRP, C-Reactive protein; ESR, Erythrocyte sedimentation rate; LDH, Lactate dehydrogenase; NLR, Neutrophil-lymphocyte ratio; NEWS 2, National Early Warning score 2; q SOFA, Quick SOFA score; SaO 2/FiO2, Saturation/Fraction of inspired oxygen.
The most common comorbidities were type 2 diabetes mellitus and hypertension with higher prevalence (2.6 and 1.8 times higher respectively) in hospitalized than outpatient management, Table 1. Other comorbidities were chronic obstructive pulmonary disease (COPD) in 4 patients, ischemic cardiomyopathy in 2 patients and vitiligo in 1 patient.
There were statistically significant changes (p < 0.001) with a 3.6 times higher NEWS 2 and 11-fold increased qSOFA score as well as decreased SaO2/FiO2 ratio in hospitalized versus outpatient management Table 1.
Hospitalized patients also showed higher values of leukocytes (2.1-fold), neutrophil/lymphocyte ratio (2.2-fold), CRP (5.3-fold), LDH (1.76-fold), ESR (1.6-fold), Ferritin (2.2-fold), D-dimer (3.5-fold) and fibrinogen (1.3-fold). All these differences were statistically significant (p < 0.01) Table 1.
Regarding the lipid profile, they showed 25% lower HDL c blood levels (p < 0.01) and increased Tg/HDL c (2.7-fold) ratio than those of outpatient management (p < 0.01) Table 2 .
Table 2.
Lipid profile of patients with COVID-19.
Variable | Global (n = 43) |
Outpatient management (n = 14) |
Hospitalized (n = 29) |
p |
---|---|---|---|---|
Median (range) | ||||
Triglyceride (mg/dL) | 147 (72–513) | 138 (83–231) | 164 (72–513) | 0.159 |
Total cholesterol (mg/dL) | 142.5 (70–283) | 147 (70–192) | 130 (71–283) | 0.554 |
HDL c (mg/dL) | 29.5 (9.6–63.5) | 40.5 (20.1–51.6) | 25.6 (9.6–63.5) | 0.009 |
LDL c (mg/dL) | 76.4 (25–166) | 68 (28.1–120) | 78 (25–166) | 0.822 |
Tg/HDL c | 5 (1.3–26.4) | 3.3 (2.7–7.2) | 6.4 (1.3–26.4) | 0.005 |
Table 2: The value of p was derived from ∗U Mann Whitney test. Statistical significance p values are shown in bold (p < 0.05).
Abbreviations: HDL c, High-density lipoprotein cholesterol; LDL c, Low-density lipoprotein cholesterol; Tg/HDL c, Triglyceride/High-density lipoprotein cholesterol ratio.
These differences were also observed between the critical care COVID-19 patients and the ones with mild symptoms, Games-Howell post-hoc test indicate significant lower HDL c (p < 0.05) and higher Tg/HDL c ratio (p < 0.05) levels Table 3 .
Table 3.
Differences in the lipid profile of patients with COVID-19 by degree of severity.
Variable | Global (n = 43) |
Mild-moderate (n = 10) |
Severe (n = 17) |
Critical (n = 16) |
p |
---|---|---|---|---|---|
Median (range) | |||||
Triglyceride (mg/dL) | 140 (72.513) | 142.5 (89–231) | 132.5 (83–288) | 167 (72–513) | 0.421 |
Total cholesterol (mg/dL) | 142.5 (70–283) | 152 (70–192) | 128 (80–283) | 136 (71–198) | 0.732 |
HDL c (mg/dL) | 29.5 (10.4–76.5) | 39.9 (20.1–51.6) | 27.6 (19.3–63.5) | 22.15 (10.4–56.5)a | 0.018 |
LDL c (mg/dL) | 76.4 (25–166) | 71.3 (28.1–120) | 78 (38.6–166) | 79.45 (25–125) | 0.903 |
Tg/HDL c | 5 (1.3–26.4) | 3.45 (2.5–7.2) | 5 (1.8–8.2) | 7.5 (1.3–26.4)a | 0.034 |
Table 3: The value of p was derived from Kruskal–Wallis test. Statistical significance p values are shown in bold (p < 0.05).
Abbreviations: HDL c, High-density lipoprotein cholesterol; LDL c, Low-density lipoprotein cholesterol; Tg/HDL c, Triglyceride/High-density lipoprotein cholesterol ratio.
Games-Howell post-hoc test (p < 0.05).
On average, hospitalization time was 15.2 ± 7.9 days. 14 patients (48.2%) had acute kidney injury, 10 (34.4%) vasopressor support and 16 (55.1%) required invasive mechanical ventilation. In the outpatient management, only 4 (28.5%) suffered acute kidney injury. 10 (34.4%) of the hospitalized patients died within the 15 days of the study.
Correlation between lipid profile, Tg/HDL c ratio and severity criteria.
A high Tg level at the time of COVID-19 diagnosis was correlated with high LDH level and NEWS 2 Table 4 . HDL c levels were correlated with SaO2/FiO2 ratio and negatively correlated with LDH, ferritin, NEWS 2, qSOFA, requirement of mechanical ventilation and days of hospitalization Table 4. Hence, the Tg/HDL c ratio were also correlated with LDH, NEWS 2, qSOFA as well as requirement of mechanical ventilation and negatively correlated with SaO2/FiO2 ratio Table 4. Total cholesterol and LDL c levels showed no correlation with the evaluated parameters Table 4.
Table 4.
Bivariate correlations between the variables studied and the lipid profile of patients with COVID-19.
Variable | Triglycerides |
Total cholesterol |
LDL c |
HDL c |
Tg/HDL c |
|||||
---|---|---|---|---|---|---|---|---|---|---|
p | r | p | r | p | r | p | r | p | r | |
CRP | 0.745 | 0.052 | 0.882 | 0.024 | 0.606 | 0.085 | 0.146 | −0.24 | 0.193 | 0.213 |
LDH | 0.050 | 0.312 | 0.323 | 0.160 | 0.066 | 0.306 | 0.049 | −0.33 | 0.030 | 0.356 |
ESR | 0.707 | 0.070 | 0.876 | −0.03 | 0.696 | −0.08 | 0.758 | −0.06 | 0.201 | 0.240 |
Ferritin | 0.540 | 0.110 | 0.400 | −0.15 | 0.817 | −0.04 | 0.009 | −0.46 | 0.113 | 0.286 |
D-dimer | 0.513 | 0.109 | 0.997 | −0.001 | 0.684 | 0.071 | 0.093 | −0.29 | 0.202 | 0.221 |
Fibrinogen | 0.633 | −0.09 | 0.998 | 0.001 | 0.941 | 0.014 | 0.907 | 0.022 | 0.473 | 0.136 |
Leucocytes | 0.158 | 0.224 | 0.780 | 0.045 | 0.922 | −0.02 | 0.380 | −0.17 | 0.171 | 0.227 |
Lymphocytes | 0.749 | −0.05 | 0.096 | −0.26 | 0.270 | −0.18 | 0.088 | −0.28 | 0.273 | 0.183 |
NLR | 0.295 | 0.168 | 0.347 | 0.151 | 0.820 | 0.038 | 0.813 | 0.04 | 0.714 | 0.061 |
SaO 2/FiO2 | 0.201 | −0.21 | 0.623 | 0.080 | 0.663 | 0.074 | 0.011 | 0.414 | 0.045 | −0.33 |
NEWS 2 | 0.043 | 0.314 | 0.952 | −0.01 | 0.991 | 0.002 | 0.002 | −0.48 | 0.001 | 0.495 |
q SOFA | 0.070 | 0.283 | 0.516 | −0.10 | 0.609 | −0.09 | <0.001 | −0.56 | <0.001 | 0.538 |
CALL score | 0.621 | 0.081 | 0.385 | 0.141 | 0.092 | 0.281 | 0.649 | 0.077 | 0.896 | −0.02 |
Acute kidney failure | 0.213 | 0.196 | 0.228 | 0.190 | 0.062 | 0.301 | 0.865 | 0.028 | 0.355 | 0.152 |
Vasopressor support | 0.254 | 0.180 | 0.717 | −0.06 | 0.270 | −0.18 | 0.209 | −0.21 | 0.324 | 0.162 |
Invasive mechanical ventilation | 0.194 | 0.204 | 0.869 | −0.03 | 0.830 | 0.036 | 0.006 | −0.43 | 0.018 | 0.378 |
Days of hospitalization | 0.542 | 0.099 | 0.419 | −0.13 | 0.683 | −0.07 | 0.031 | −0.36 | 0.064 | 0.308 |
Death | 0.476 | 0.113 | 0.908 | 0.018 | 0.295 | 0.172 | 0.119 | −0.25 | 0.167 | 0.226 |
Table 4: The value of p was derived from Spearman's Rank Correlation Coefficient. Statistical significance p values are shown in bold (p < 0.05).
Abbreviations: CALL Score, Comorbidity-Age-Lymphocyte count-Lactate dehydrogenase score; CRP, C-Reactive protein; ESR, Erythrocyte sedimentation rate; HDL c, High-density lipoprotein cholesterol; LDH, Lactate dehydrogenase; LDL c, Low-density lipoprotein cholesterol; NLR, Neutrophil-lymphocyte ratio; NEWS 2, National Early Warning score 2; q SOFA, Quick SOFA score; SaO2/FiO2, Saturation/Fraction of inspired oxygen; Tg/HDL c, Triglyceride/High-density lipoprotein cholesterol ratio.
We found a negative correlation between HDL c level and requirement of higher oxygen supply (r = −0.475; p = 0.002) and a positive correlation with the Tg/HDL c ratio (r = 0.447; p = 0.004).
Simple lineal regression analyses were used to investigate if any of the lipid profile components and/or the Tg/HDL c ratio are predictive of the biochemical variables of severity.
LDH blood levels were predicted by Tg concentration (F = 11.000; R2 = 0.225; Coefficient B = 0.90; SE 0.272; p = 0.002) and Tg/HDL c ratio (F = 12.200; R2 = 0.259; Coefficient B = 10.6; SE 3.030; p = 0.001). Ferritin blood levels were predicted by Tg concentration (F = 5.356; R2 = 0.147; Coefficient B = 2.162; SE 0.934; p = 0.027), HDLc level (F = 6.583; R2 = 0.180; Coefficient B = −20.237; SE 7.887; p = 0.016) and Tg/HDL c ratio (F = 7.929; R2 = 0.209; Coefficient B = 28.1; SE 9.980; p = 0.009). D-dimer blood levels were predicted by Tg concentration (F = 10.985; R2 = 0.234; Coefficient B = 6.2; SE 1.876; p = 0.002) and Tg/HDL c ratio (F = 16.237; R2 = 0.330; Coefficient B = 80.9; SE 20.093; p=<0.001). No association was found for CRP, ESR or fibrinogen.
Univariate logistic regression analysis also showed that predictors of invasive mechanical ventilation requirement were Tg concentration (R2 Nagelkerke = 0.173; OR 1.008, CI 1.000–1.016; p = 0.040), HDL c concentration (R2 Nagelkerke = 0.232; OR 0.919, CI 0.855–0.987; p = 0.021) and Tg/HDL c ratio (R2 Nagelkerke = 0.394; OR 1.292, CI 1.025–1.629; p = 0.03). None of the lipid profile components nor the Tg/HDL c ratio were predictors of acute kidney injury, vasopressor support or death.
The prediction for invasive mechanical ventilation requirement was also determined by ROC curve analysis calculating the AUC. HDL c levels and Tg/HDL c ratio predicted requirement of invasive mechanical ventilation Fig. 1 .
Fig. 1.
Analysis of the ROC curve of the triglyceride level, HDL c and the Tg/HDL c ratio for the requirement of invasive mechanical ventilation in patients with COVID-19. Fig. 1: The value of p was derived from ∗ Chi squared test, ∗∗ Fisher's exact test. Statistical significance p < 0.05. Abbreviations: AUC, Area under the curve; CI, Confidence interval; HDL c, High-density lipoprotein cholesterol; ROC, Receiver-Operating Characteristic; TG, Triglyceride; Tg/HDL c, Triglyceride/High-density lipoprotein cholesterol ratio.
Determination of cut-off points based on the maximum value of the Youden Index revealed that ≤35 mg/dL of HDL c blood concentration showed a relative risk (RR) for the requirement of invasive mechanical ventilation Fig. 1 and to have a SaO2/FiO2 index <160 Fig. 2 . For the Tg/HDL c ratio a value of ≥7.45 had a RR of 3.84 for requirement of invasive mechanical ventilation Fig. 1 and RR of 3.69 to have a SaO2/FiO2 index <160 Fig. 2.
Fig. 2.
Analysis of the ROC curve of the HDL c and the Tg/HDL c ratio with a SaO2/FiO2 index <160 in patients with COVID-19. Fig. 2: The value of p was derived from ∗ Chi squared test, ∗∗ Fisher's exact test. Statistical significance p < 0.05. Abbreviations: AUC, Area under the curve; CI, Confidence interval; HDL c, High-density lipoprotein cholesterol; ROC, Receiver-Operating Characteristic; SaO2/FiO2, Saturation/Fraction of inspired oxygen; Tg/HDL c, Triglyceride/High-density lipoprotein cholesterol ratio.
HDL c blood concentration ≤35 mg/dL and Tg/HDL c ratio ≥7.45 also predicted a NEWS 2 ≥ 2 Fig. 3 .
Fig. 3.
Analysis of the ROC curve of the HDL c and the Tg/HDL c ratio with NEWS 2 in patients with COVID-19. Fig. 3: The value of p was derived from ROC analysis. Statistical significance p < 0.05. Abbreviations: AUC, Area under the curve; CI, Confidence interval; HDL c, High-density lipoprotein cholesterol; NEWS 2, National Early Warning score 2; ROC, Receiver-Operating Characteristic; Tg/HDL c, Triglyceride/High-density lipoprotein cholesterol ratio.
Binomial multivariate logistic regression revealed that ≤35 mg/dL HDL c blood concentration and ≥7.45 Tg/HDL c ratio are predictors adjusted by age and gender for the requirement of invasive mechanical ventilation Table 5 .
Table 5.
HDL c level and Tg/HDL c ratio as risk factors adjusted to age and gender for the requirement of invasive mechanical ventilation in patients with COVID-19.
Variable | Sig Chi2 | R2 Nagelkerke | Coefficient B | SE | OR Multivariate (CI 95%) | p | |
---|---|---|---|---|---|---|---|
Step 1 | HDL c ≤ 35 mg/dL | 2.619 | 1.134 | 13.716 (1.485–126.654) | 0.021 | ||
Tg/HDL c ≥ 7.45 | <0.001 | 0.542 | 2.469 | 0.951 | 11.815 (1.832–76.186) | 0.009 | |
Age | −0.064 | 0.035 | 0.938 (0.876–1.005) | 0.069 | |||
Gender | −0.246 | 0.873 | 0.782 (0.141–4.325) | 0.778 | |||
Constant | 1.858 | 2.220 | 6.409 | 0.403 | |||
Step 2 | HDL c ≤ 35 mg/dL | 2.633 | 1.129 | 13.916 (1.523–127.171) | 0.020 | ||
Tg/HDL c ≥ 7.45 | <0.001 | 0.541 | 2.463 | 0.952 | 11.742 (1.817–75.854) | 0.010 | |
Age | −0.064 | 0.035 | 0.938 (0.876–1.006) | 0.071 | |||
Constant | 1.694 | 2.162 | 5.440 | 0.433 |
Table 5: The value of p was derived from Multivariate logistic regression analysis with successive steps backward (conditional) method. Statistical significance p values are shown in bold (p < 0.05).
Abbreviations: HDL c, High-density lipoprotein cholesterol; LDL c, Low-density lipoprotein cholesterol; OR, odds ratio; SD, standard deviation; Tg/HDL c, Triglyceride/High-density lipoprotein cholesterol ratio.
4. Discussion
Key factors linked to the development of cardiovascular disease, hypertension and type 2 diabetes mellitus are obesity, dyslipidemia and insulin resistance [13].
Mexico has a high prevalence of metabolic syndrome and related chronic diseases. According to the most recent national health and nutrition survey (Encuesta Nacional de Salud y Nutrición [ENSANUT] de Medio Camino 2016) [14], Mexican adults have a 71.2% prevalence of overweight and obesity, 25.5% have hypertension and 13.7% have type 2 diabetes mellitus [14,15]. In ENSANUT 2012, was reported dyslipidaemia in 50- to 59-year-olds with a 36.8% prevalence of ≥200 mg/dL total cholesterol; 58.7% of <40 mg/dL HDL c; 60.1% of ≥150 mg/dl Tg and 61.7% of ≥100 mg/dL LDL c. A common phenotype in this age group was that of high Tg with low HDL c blood levels [16].
In Mexico, by February 26th 2021, there was 2,069,370 confirmed cases and 183,692 associated COVID-19 deaths becoming the country with the highest mortality rate (8.9%) with 145.57 deaths per 100,000 inhabitants [17].
This high mortality rate in COVID-19 patients may be related with the elevated risk to develop severe complications in individuals with cardiovascular disease, hypertension and type 2 diabetes mellitus [8]. As dyslipidaemia is a common feature in these metabolic disorders and has been linked with severe COVID-19 [9,10]; our results confirm that dyslipidaemia and particularly a high Tg/HDL c ratio at the time of COVID-19 diagnosis can be used as a prognosis factor for disease severity and requirement for invasive mechanical ventilation.
Although we found no differences in gender, age or comorbidities between hospitalized versus outpatient management, a higher percentage of men 75%, required hospitalization compared with 57.8% of women; a point already observed for COVID-19 complications and viremia [8,18,19]. Additionally, 72% of patients with type 2 diabetes mellitus and 64.7% with hypertension were hospitalized. These are commonly reported comorbidities in hospitalized COVID-19 patients [8,18,19].
Our analysis revealed that a low plasmatic level of HDL c and elevated Tg/HDL c ratio at the time of COVID-19 diagnosis is linked with requirement of hospitalization and illness severity as critical care patients showed the lowest HDL c concentration and highest Tg/HDL c ratio compared with severe and mild/asymptomatic COVID-19 patients.
Elevated Tg and low HDL c were constantly associated with clinical and biochemical data of COVID-19 severity like increased LDH levels which is an inflammatory marker previously proposed as predictor of poor prognosis in COVID-19 patients [20,21] and with NEWS 2 an early warning score published by the Royal College of Physicians [22] with a good performance in the prediction of severity and mortality in COVID-19 patients [[22], [23], [24]]. A low HDL c concentration was additionally linked with high ferritin levels, a protein associated with inflammatory processes and previously reported increased in severe COVID-19 patients [25], related with SARS-CoV-2 viremia [18] and proposed as predictor of mortality for the disease [26,27]. In addition, low HDL c levels were associated with high qSOFA, another early warning score associated with COVID-19 mortality [27]. Remarkably, we found negative correlations between plasmatic HDL c and length of hospitalization along with requirement of invasive mechanical ventilation and a positive correlation with SaO2/FiO2 index, all variables related to the severity of COVID-19.
Our results are in line with those reported previously [[28], [29], [30]] of a higher prevalence of severe COVID-19 cases in patients with low HDL c levels. There is also observed an association between lower concentrations of apolipoprotein ApoA 1, one of its major structural components, which is inversely correlated with inflammatory states, disease severity and mortality [30,31].
HDL c has immunomodulatory effects via its binding to pathogen-associated molecules like bacterial-lipopeptides (lipopolysaccharide, lipoteichoic acid) [32] and diacylated peptides like Pam2CSK4, FSL-1 which neutralize the infectious activity by blocking TLR2 and TLR4 responses [33] as well as by inhibiting the production of inflammatory cytokines by macrophages [34].
Besides, elevated serum HDL c can also promote proliferation of IgA in early stages of bacterial infection [35], but it is not known if a similar mechanism could be prevented COVID-19 viremia and derived complications. Nevertheless, as HDL c particles transport paraoxonase 1 which has antiviral properties; this could induce virus inactivation [36] and ApoA 1 has shown protective effects in several lung disease conditions, including viral pneumonia [37] although the molecular mechanism involved is not yet elucidated.
Given the observed associations of clinical and biochemical markers of severity with high Tg and low HDL c, our analysis of Tg/HDL c ratio correlations reveal positive associations with LDH, NEWS 2, qSOFA, length of hospitalization, requirement of invasive mechanical ventilation and a positive correlation with SaO2/FiO2 index.
To our knowledge this is the first study assessing the correlation between Tg/HDL c ratio and COVID-19 complications.
Tg/HDL c ratio has been considered as a better marker of cardiovascular risk than serum concentrations of HDL c and LDL c alone [12] and given its good correlation with the hyperinsulinemic-euglycemic clamp, an accurate estimation of insulin sensitivity; Tg/HDL c ratio can also reflect glucose-metabolic alterations [38].
Hyperinsulinemia of patients with insulin resistance and diabetes could also contribute to increased SARS-CoV-2 viremia [18] as insulin increases membrane expression of angiotensin-converting enzyme 2 (ACE 2) in pneumocyte [39] which function as receptor of the SARS-CoV-2 spike protein causing the cell infection.
In our study, higher Tg/HDL c ratio were observed in hospitalized versus outpatient management and in critical care patients versus mild/asymptomatic individuals. Additionally, high Tg/HDL c ratio was also correlated with increased need for oxygen support.
Greater risk to develop complications and requirement of oxygen assistance are reported in COVID-19 patients with diabetes [18,19]. Hyperinsulinaemia and hyperglycaemia increase inflammation and risk of thrombosis by increasing coagulation [40]. Hyperinsulinemia increases plasminogen activator type 1 levels, promoting thrombi by inhibition of fibrinolysis while hyperglycaemia increase blood coagulation and production of proinflammatory cytokines TNF-alpha and IL-6 [41].
Pulmonary thrombi may contribute to oxygen desaturation and respiratory distress in COVID-19 cases [40] and this could be promoted by insulin resistance which raise glucose and insulin circulating levels.
However, more studies are needed to associate hyperinsulinemia with inflammatory and thrombotic processes in COVID-19, and its direct link with Tg/HDL c ratio in severe patients could indicate hypoxemia and deficient pulmonary function in COVID-19 severe and critical patients.
5. Conclusion
Our results emphasize the use of parameters of the lipid profile such as HDL c level as a marker of risk for severe COVID-19 outcomes. Moreover, our findings disclose that the Tg/HDL c ratio is a novel biochemical marker of severe prognosis and requirement of invasive mechanical ventilation in COVID-19 patients.
6. Limitations
One of the limitations of the present study is the lack of information about the lipid profile of the patients before infection. There are reports about effects of SARS-CoV-2 on metabolic features and one observation is decreased levels of total cholesterol, LDL c and HDL c, that fell continuously until the 9th day of infection and then return to normal levels [36,42]. Therefore, it is not possible to exclude a viral effect on the low HDL c levels found in severe COVID-19 patients at the time of diagnosis. Another limitation is the reduced sample size; however, the strength is the surveillance of the patients during the 15 days following COVID-19 diagnosis, with a very similar treatment scheme for an accurate variable association in a homogeneous population.
7. Statement of authorship
AAE: Conceptualization, data collection, statistical analysis, data interpretation and article writing. MPE: Conceptualization, data interpretation and article writing. MRF: Conceptualization. BMDR: Conceptualization. GLJA: Data collection. MPMA: Data collection. LNJJ: statistical analysis. AAV: data interpretation and article writing. All authors read and approved the final manuscript.
Funding source
Secretaría de Salud de la Ciudad de México (SEDESA).
Declaration of competing interest
The authors declare no competing interests.
References
- 1.di Mauro Gabriella, Cristina S., Concetta R., Francesco R., Annalisa C. SARS-Cov-2 infection: response of human immune system and possible implications for the rapid test and treatment. Int Immunopharm. 2020;84 doi: 10.1016/j.intimp.2020.106519. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bulut C., Kato Y. Epidemiology of COVID-19 (special issue: COVID-19 pandemic) Turk J Med Sci. 2020:50. doi: 10.3906/sag-2004-172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ye Q., Wang B., Mao J. The pathogenesis and treatment of the ‘Cytokine Storm’’ in COVID-19. J Infect. 2020;80 doi: 10.1016/j.jinf.2020.03.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Oliveira D.S., Medeiros N.I., Gomes J.A.S. Immune response in COVID-19: what do we currently know? Microb Pathog. 2020;148 doi: 10.1016/j.micpath.2020.104484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ifrim D.C., Quintin J., Joosten L.A.B., Jacobs C., Jansen T., Jacobs L., et al. Trained immunity or tolerance: opposing functional programs induced in human monocytes after engagement of various pattern recognition receptors. Clin Vaccine Immunol. 2014;21 doi: 10.1128/CVI.00688-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Jialal I., Kaur H., Devaraj S. Toll-like receptor status in obesity and metabolic syndrome: a translational perspective. J Clin Endocrinol Metab. 2014;99 doi: 10.1210/jc.2013-3092. [DOI] [PubMed] [Google Scholar]
- 7.Jialal I., Huet B.A., Kaur H., Chien A., Devaraj S. Increased toll-like receptor activity in patients with metabolic syndrome. Diabetes Care. 2012;35 doi: 10.2337/dc11-2375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zheng Z., Peng F., Xu B., Zhao J., Liu H., Peng J., et al. Risk factors of critical & mortal COVID-19 cases: a systematic literature review and meta-analysis. J Infect. 2020:81. doi: 10.1016/j.jinf.2020.04.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Choi G.J., Kim H.M., Kang H. The potential role of dyslipidemia in covid-19 severity: an umbrella review of systematic reviews. J Lipid and Atherosclerosis. 2020;9 doi: 10.12997/jla.2020.9.3.435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hariyanto T.I., Kurniawan A. Dyslipidemia is associated with severe coronavirus disease 2019 (COVID-19) infection. Diabetes and Metabolic Syndrome: Clin Res Rev. 2020;14 doi: 10.1016/j.dsx.2020.07.054. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Baez-Duarte Bg Z-Gig-Drt-Rer-Vgp-Frctmrgod Triglyceride/high-density lipoprotein cholesterol (TG/HDL-C) index as a reference criterion of risk for metabolic syndrome (MetS) and low insulin sensitivity in apparently healthy subjects. Gac Med Mex. 2017;153:152–158. [PubMed] [Google Scholar]
- 12.Yang S.H., Du Y., Li X.L., Zhang Y., Li S., Xu R.X., et al. Triglyceride to high-density lipoprotein cholesterol ratio and cardiovascular events in diabetics with coronary artery disease. Am J Med Sci. 2017;354 doi: 10.1016/j.amjms.2017.03.032. [DOI] [PubMed] [Google Scholar]
- 13.Ormazabal V., Nair S., Elfeky O., Aguayo C., Salomon C., Zuñiga F.A. Association between insulin resistance and the development of cardiovascular disease. Cardiovasc Diabetol. 2018;17 doi: 10.1186/s12933-018-0762-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.ENSANUT-MC. 2016. https://ensanut.insp.mx/encuestas/ensanut2016/index.php. n.d (cited 26 Feb 21) Available from: [Google Scholar]
- 15.Basto-Abreu A., Barrientos-Gutiérrez T., Rojas-Martínez R., Aguilar-Salinas C.A., López-Olmedo N., de la Cruz-Góngora V., et al. Prevalencia de diabetes y descontrol glucemico en Mexico: resultados de la Ensanut 2016. Salud Publica Mex. 2020;62 doi: 10.21149/10752. [DOI] [PubMed] [Google Scholar]
- 16.Hernández-Alcaraz C., Aguilar-Salinas C.A., Mendoza-Herrera K., Pedroza-Tobías A., Villalpando S., Shamah-Levy T., et al. Dyslipidemia prevalence, awareness, treatment and control in Mexico: results of the Ensanut 2012. Salud Publica Mex. 2020;62 doi: 10.21149/10520. [DOI] [PubMed] [Google Scholar]
- 17.Mortality Analyses Coronavirus resource center. Johns Hopkins University and Medicine. 2021 https://coronavirus.jhu.edu/data/mortality n.d (cited 26 Feb 2021) Available from: [Google Scholar]
- 18.Myhre P.L., Prebensen C., Monceyron Jonassen C., Erik Berdal J., Omland T. SARS-CoV-2 viremia is associated with inflammatory, but not cardiovascular biomarkers, in patients hospitalized for COVID-19. J Am Heart Ass. 2021;10(9) doi: 10.1161/JAHA.120.019756. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Chaudhary R., Garg J., Houghton D.E., Murad M.H., Kondur A., Chaudhary R., et al. Thrombo-inflammatory biomarkers in COVID-19: systematic review and meta-analysis of 17,052 patients. Mayo Clin Proc: Innovations, Quality & Outcomes. 2021;5(2):388–402. doi: 10.1016/j.mayocpiqo.2021.01.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pan F., Yang L., Li Y., Liang B., Li L., Ye T., et al. Factors associated with death outcome in patients with severe coronavirus disease-19 (Covid-19): a case-control study. Int J Med Sci. 2020;17 doi: 10.7150/ijms.46614. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Rojas-Marte G., Hashmi A.T., Khalid M., Chukwuka N., Fogel J., Munoz-Martinez A., et al. Outcomes in patients with COVID-19 disease and high oxygen requirements. J Clin Med Res. 2021:13. doi: 10.14740/jocmr4405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Baker K.F., Hanrath A.T., van der Loeff I.S., Kay L.J., Back J., Duncan C.J. National Early Warning Score 2 (NEWS2) to identify inpatient COVID-19 deterioration: a retrospective analysis. Clin Med. 2021;21(2):84–89. doi: 10.7861/clinmed.2020-0688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Myrstad M., Ihle-Hansen H., Tveita A.A., Andersen E.L., Nygård S., Tveit A., et al. National Early Warning Score 2 (NEWS2) on admission predicts severe disease and in-hospital mortality from Covid-19 - a prospective cohort study. Scand J Trauma Resuscitation Emerg Med. 2020;28 doi: 10.1186/s13049-020-00764-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kostakis I., Smith G.B., Prytherch D., Meredith P., Price C., Chauhan A., et al. The performance of the National Early Warning Score and National Early Warning Score 2 in hospitalised patients infected by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Resuscitation. 2021;159 doi: 10.1016/j.resuscitation.2020.10.039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Vargas-Vargas M., Cortés-Rojo C. Ferritin levels and COVID-19. Revista Panamericana de Salud Publica/Pan American Journal of Public Health. 2020;44 doi: 10.26633/RPSP.2020.72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Ahmed S., Ansar Ahmed Z., Siddiqui I., Haroon Rashid N., Mansoor M., Jafri L. Evaluation of serum ferritin for prediction of severity and mortality in COVID-19- A cross sectional study. Annals of Med and Surg. 2021;63 doi: 10.1016/j.amsu.2021.02.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Guo W., Ran L., Zhu J., Ge Q., Du Z., Wang F., et al. Identifying critically ill patients at risk of death from coronavirus disease. World J Emergency Med. 2021;12 doi: 10.5847/wjem.j.1920-8642.2021.01.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Wang G., Zhang Q., Zhao X., Dong H., Wu C., Wu F., et al. Low high-density lipoprotein level is correlated with the severity of COVID-19 patients: an observational study. Lipids Health Dis. 2020;19 doi: 10.1186/s12944-020-01382-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Peng Y., Wan L., Fan C., Zhang P., Wang X., Sun J., et al. Cholesterol metabolism—impacts on SARS-CoV-2 infection prognosis. MedRxiv. 2020 doi: 10.1101/2020.04.16.20068528. https://www.medrxiv.org/content/10.1101/2020.04.16.20068528v3 [DOI] [Google Scholar]
- 30.Nie S., Zhao X., Zhao K., Zhang Z., Zhang Z., Zhang Z. Metabolic disturbances and inflammatory dysfunction predict severity of coronavirus disease 2019 (COVID-19): a retrospective study. MedRxiv. 2020 doi: 10.1101/2020.03.24.20042283. https://www.medrxiv.org/content/10.1101/2020.03.24.20042283v1 [DOI] [Google Scholar]
- 31.Sun J.T., Chen Z., Nie P., Ge H., Shen L., Yang F., et al. Lipid profile features and their associations with disease severity and mortality in patients with COVID-19. Frontiers in Cardiovasc Med. 2020;7 doi: 10.3389/fcvm.2020.584987. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Sorokin A.v., Karathanasis S.K., Yang Z.H., Freeman L., Kotani K., Remaley A.T. COVID-19—associated dyslipidemia: implications for mechanism of impaired resolution and novel therapeutic approaches. FASEB (Fed Am Soc Exp Biol) J. 2020;34 doi: 10.1096/fj.202001451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.van Bergenhenegouwen J., Kraneveld A.D., Rutten L., Garssen J., Vos A.P., Hartog A. Lipoproteins attenuate TLR2 and TLR4 activation by bacteria and bacterial ligands with differences in affinity and kinetics. BMC Immunol. 2016;17 doi: 10.1186/s12865-016-0180-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.de Nardo D., Labzin L.I., Kono H., Seki R., Schmidt S.v., Beyer M., et al. High-density lipoprotein mediates anti-inflammatory reprogramming of macrophages via the transcriptional regulator ATF3. Nat Immunol. 2014;15 doi: 10.1038/ni.2784. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Niki M., Yoshiyama T., Nagai H., Miyamoto Y., Niki M., Oinuma K., et al. Nutritional status positively impacts humoral immunity against its Mycobacterium tuberculosis, disease progression, and vaccine development. PloS One. 2020;15 doi: 10.1371/journal.pone.0237062. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Kočar E., Režen T., Rozman D. Cholesterol, lipoproteins, and COVID-19: basic concepts and clinical applications. Biochim Biophys Acta Mol Cell Biol Lipids. 2021:1866. doi: 10.1016/j.bbalip.2020.158849. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Gordon E.M., Figueroa D.M., Barochia A.v., Yao X., Levine S.J. High-density lipoproteins and apolipoprotein A-I: potential new players in the prevention and treatment of lung disease. Front Pharmacol. 2016;7 doi: 10.3389/fphar.2016.00323. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Matsuda M., DeFronzo R.A. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care. 1999;22 doi: 10.2337/diacare.22.9.1462. [DOI] [PubMed] [Google Scholar]
- 39.Finucane F.M., Davenport C. Coronavirus and obesity: could insulin resistance mediate the severity of covid-19 infection? Frontiers in Public Health. 2020;8 doi: 10.3389/fpubh.2020.00184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Cooper I.D., Crofts C.A.P., DiNicolantonio J.J., Malhotra A., Elliott B., Kyriakidou Y., et al. Relationships between hyperinsulinaemia, magnesium, vitamin D, thrombosis and COVID-19: rationale for clinical management. Open Heart. 2020;7 doi: 10.1136/openhrt-2020-001356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Stegenga M.E., van der Crabben S.N., Levi M., de Vos A.F., Tanck M.W., Sauerwein H.P., et al. Hyperglycemia stimulates coagulation, whereas hyperinsulinemia impairs fibrinolysis in healthy humans. Diabetes. 2006;55 doi: 10.2337/db05-1543. [DOI] [PubMed] [Google Scholar]
- 42.Hu X., Chen D., Wu L., He G., Ye W. Low serum cholesterol level among patients with COVID-19 infection in wenzhou, China. SSRN Electronic J. 2020 doi: 10.2139/ssrn.3544826. https://ssrn.com/abstract=3544826 [DOI] [Google Scholar]