Serum lipid profile changes and their clinical diagnostic significance in COVID-19 Mexican Patients

1. Introduction

Coronavirus disease 2019 (COVID-19) is a public health emergency of international concern caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). More than two hundred countries around the world have reported more than 13 million confirmed cases, and more than half a million deaths caused by COVID-19^1,2. The epidemiological information of the COVID-19 pandemic in Latin America indicates that as of June nearly four million cases and more than two hundred thousand deaths have been reported³, while in Mexico, 311, 486 cases of COVID-19 and 36,327 deaths have been reported^4,5. The SARS-CoV-2 transmission was initially linked to a local exotic animal market in the province of Wuhan, China, indicating a possible zoonosis ⁶ SARS-CoV-2 can be transmitted human-to-human, through the inhalation of aerosols with nasal discharge droplets, in addition to direct or indirect contact of the oral and nasal mucous membranes and eye membranes with everyday contaminated objects ^7-10 The symptoms of the disease may be mild (flu-like infection with fever, headache or dry cough) or severe due to acute lung injury (ALI) and acute respiratory distress syndrome (ARDS), leading to acute breathing difficulties, respiratory failure, multiple organ failure, or death of the patient ^9-11. Reports of patients with associated morbidity, mainly diabetes (DM), hypertension (HT), and other cardiovascular and cerebrovascular diseases, have been linked to poor prognosis of the disease^12-14.

Additionally, manifestations of hypolipidemia were reported in Chinese patients with mild symptoms and progressively worsened according to the severity of the disease^15,16. The severity of the disease could be attributed to alterations in cholesterol levels since mainly two groups of the population are more affected in morbidity and mortality by COVID-19^17-21. On the other hand, changes in serum lipid profiles have been reported previously in other viral infections^22-26, including respiratory viral infections, where their potential as metabolic biomarkers to help clinicians distinguish bacterial from viral infection in febrile patients has been determined²⁷. For these reasons, we found it interesting to explore whether alterations in serum lipid levels in COVID-19 Mexican patients can be useful as biomarkers of disease or severity. In this study, we analyzed the lipid profile of one hundred two COVID-19 positive patients from 3 hospitals in Culiacan, Sinaloa, in northwest Mexico, one of the states with the highest number of cases of the disease in Mexico, and using a combination of ROC curves and logistic regression analysis we evaluate the clinical diagnosis utility of the lipid profile in Mexican COVID-19 patients.

1. Material and methods

3. Results

During the study period, clinical and lipid profile information was obtained from 102 COVID-19 patients. Their demographic data showed a median age of 57 years and a higher frequency of male patients at hospital admission (Table 1). The patient stratification was according to the risk, based on the clinic, as shown in Table 1. Age differs significantly between non-critical and critical COVID-19 cases (p= 0.0095), and males were more frequent among non-critical and critical COVID-19 patients. About comorbidities, hypertension (HT), obesity, and diabetes (DM) were the most frequent, especially in the critical COVID-19 group of patients (Table 1).

Regarding to the lipid profile in COVID-19 patients, the mean cholesterol (126 ± 41.02 mg/dL), LDL (69.5 ± 34.9 mg/dL) and HDL (29.9 ± 12.19 mg/dL) levels were lower compared to healthy control group (157.7 ± 37.1 mg/dL, LDL: 100.6 ± 27.64mg/dL and 41.43 ± 15.01 mg/dL, respectively) (Figure 1). On the other hand, both mean triglyceride (155.4 ± 72.49 mg/dL) and VLDL (31.09 ± 14.5 mg/dL) levels were higher in COVID-19 patients compared to healthy control (115 ± 65.07 mg/dL and 23 ± 13.01 mg/dL) (Figure 1).

• Download figure
• Open in new tab

Figure 1.

Lipid profile in COVID-19 patients (black circles) compared with healthy controls (Gray circles). In blue is showed the mean line and SD error bars lines. The output analysis is represented as the following p-values means differences: 0.0332 (*), 0.0021 (**), 0.0002 (***), <0.0001 (****).

To establish if differences between the lipid profile can accurately use to distinguish between COVID-19 patients and healthy individuals, ROC curves were used to determine optimal cut-off points and AUC for each lipid parameter. Based on Youden’s J index, the results of the ROC analysis, optimal cut-off, and the AUC calculations are shown in Table 2. Cholesterol and LDL showed optimal predictive value as well as higher sensitivity (greater than 60%) and specificity (greater than 80%) and AUC values of 0.710 and 0.769, respectively (Table 2).

In further analysis, to determine if there were differences between the lipid profile in COVID-19 patients according to severity, non-critical and critical. Lower cholesterol (119 mg/dL, IQR: 94.5 - 154.5 mg/dL, n= 38), LDL (50 mg/dL, IQR: 33 - 87.75 mg/dL, n= 36), and HDL (26 mg/dL, IQR: 16-43 mg/dL, n= 35) values were observed in critical compared with non-critical patients (cholesterol: 125 mg/dL, IQR: 102 - 160.8, n= 64; LDL: 69 mg/dL, IQR: 48.25 - 95.25 mg/dL, n= 58; HDL: 30 mg/dL, IQR: 22 - 36.5 mg/dL, n= 61), while median values of triglycerides (145 mg/dL, IQR: 100.5 - 201.5 mg/dL, n= 37) and VLDL in critical patients were higher than median triglycerides and VLDL values in no-critical (138 mg/dL, IQR: 103 – 105 mg/dL, n= 61, and 27.6 mg/dL, IQR: 20.6 - 41 mg/dL, for triglycerides and VLDL, respectively). However, only median LDL values differences were statistically significant among non-critical and critical COVID-19 patients (Figure 2).

• Download figure
• Open in new tab

Figure 2.

Lipid profile compared between no-critical (gray) and critical (black) COVID-19 patients. In blue is showed the median line and IQR error bars lines. The output analysis is represented as the following p-values mean differences: 0.0332 (*).

Moreover, a combination of ROC curves and binary logistic regression analysis was performed as a predictive model. The optimal working point was for LDL and showed predictive values of 50% sensitivity and 75% specificity, as well as an AUC of 0.621, with an optimal cut-off value ≤ 48 mg/dL, to discriminate between non-critical and critical COVID-19 (Table 3).

On the other hand, the binary logistic regression analysis on the correlations showed that LDL ≤ 48 mg/dL cut-off value might be a risk factor for critical COVID-19 (OR= 2.07, 95% IC: 1.18 to 3.63; p= 0.01) (Table 4).

4. Discussion

Lipids are essential components of the human body, and they are involved in metabolic diseases such as diabetes, which is characterized by elevated total cholesterol, LDL, HDL, and triglycerides serum levels ^31,32. Also, increased serum levels of cholesterol, triglycerides, LDL and decreased HDL are known to be associated with the significant risk factors for Cardiovascular Disease and hypertension³³. Interestingly, changes in serum lipid profiles are prognostic indicators of diseases, including viral infections. In this sense, clinical research indicates that patients infected with human immunodeficiency (HIV)²², dengue (DENV) ^23-25, and hepatitis B (HBV) ²⁶ viruses develop hypocholesterolemia and hypertriglyceridemia, as well as low levels of LDL or HDL.

A recently published study performed in China found that LDL and total cholesterol levels were significantly lower in COVID-19 patients as compared with the control group ¹⁵. In the same way of evidence, Fan et al. reported a similar pattern of low cholesterol, LDL, and HDL levels in a group of 21 patients with COVID-19 in Wuhan, China ¹⁶ Therefore, it appears that there is a pattern of dyslipidemia that may be characteristic of patients with COVID-19. However, it is true that worldwide there is a wide variation in serum lipid profile levels depending on the different population groups³³.

In the present study, conducted in a Mexican population, significant changes in the lipid profile of patients with COVID-19 was found. The same pattern of dyslipidemia observed in previously reported studies in Asia was also observed in the Mexican population ^15,16, as low levels of cholesterol, LDL, and HDL, while high levels of triglycerides and VLDL when compared with a group of healthy individuals. This repeating pattern in lipid profile changes in COVID-19 patients made us propose the lipid profile changes as a possible biomarker for COVID-19. Therefore, through a predictive model using ROC curves, we were able to detect the optimal cut-off points of ≤99 mg/dL for cholesterol and ≤48 mg/dL for LDL, which could be considered acceptable predictors for COVID-19 since they demonstrated to have significant AUC values between 0.70 and 0.80 for diagnostic accuracy ³⁴.

Serum dyslipidemia during the COVID-19 may be due to the role of lipids in SARS-Cov-2 replication in the host cell since they participate in the internalization of the virus into the cell^35–37, and in the dynamism of the lipid rafts in the membrane where the ACE-2 receptor is located ^38,39. Also, SARS-CoV-2 regulates the lipidic metabolism to generate viral membrane and pack its genome and in the exocytosis of the virions ^40,41. For this reason, a growing recent interest in the lipids changes research has been noted in SARS-CoV-2 infected patients, since people with metabolic diseases such as DM, HT, and obesity, possibly leading to a higher risk of complications and mortality from COVID-19 ^42,43 Our study found high frequencies of comorbidities alone or in combination, such as HT, DM, and obesity in COVID-19 patients. One of the main concerns of the Institute of Public Health is the high prevalence of these comorbidities in the Mexican population ^44,45.

At this moment, COVID-19 is considering as an emerging and rapidly evolving condition. So, efforts to determine the change in laboratory parameters as biomarkers have been directed towards the identification of severe or critical cases of COVID-19 ^46,47 In this regard, we stratified into non-critical and critical patients, and the same pattern of low levels of cholesterol, LDL, and HDL was observed in critical COVID-19 patients.

The above is in concordance with observed by Wei et al. previously reported¹⁵. They divided COVID-19 patients per their severity into mild, severe, and critical categories. They found that the development of hypolipidemia begins in patients with mild symptoms and progressively worsens in association with the severity of the disease ¹⁵ However, they reported that triglyceride levels decreased significantly in critical cases compared to mild and severe cases. Contrary to our results, as higher triglyceride levels were found in critically ill patients with COVID-19, but only median LDL levels differences were statistically different between non-critical and critical patients. In another similar research, Fan et al., besides reporting a pattern of dyslipidemia similar to what has previously been observed in patients with COVID-19, they also showed that degrees of decrease in LDL levels are highly likely to be associated with the severity and mortality of the disease¹⁶. The optimal cut-off point was for LDL in the ROC analysis, and we observed that LDL might be a risk factor for critical COVID-19. Because of the low AUC value obtained for LDL, it could not be considered an acceptable biomarker to differentiate between non-critical and critical COVID-19 patients. This may be due to the limitations of our study, such as sample size or the low statistical power difference observed between both groups of patients. However, it cannot be ruled out that LDL levels may be useful for decision making in the evolution of COVID-19 critically ill patients, and more studies are needed, especially in high-risk populations.

Funding

None.

Declaration of Competing Interest

None.