Abstract
Background
Triglyceride-glucose (TyG) index is a simple and reliable surrogate marker for insulin resistance. Epidemiology studies have shown that insulin resistance is a risk factor for various infectious diseases. We evaluated the prognostic value of TyG index measured before the COVID-19 infection in COVID-19 infected patients.
Methods
From a nationwide COVID-19 cohort dataset in Korea, we included COVID-19 patients diagnosed between Jan and Jun 2020. Based on the nationwide health screening data between 2015 and 2018, TyG index was calculated as ln [triglyceride (mg/dL) × fasting glucose level (mg/dL)/2]. Primary outcome is development of severe complications of COVID-19 defined as composite of mechanical ventilation, intensive care unit care, high-flow oxygen therapy, and mortality within two months after the diagnosis of COVID-19.
Results
This study included 3887 patients with COVID-19 confirmed by reverse transcription polymerase chain reaction. Mean ± standard deviation of TyG index was 8.54 ± 0.61. Severe complications of COVID-19 were noted in 289 (7.44%) patients. In the multivariate logistic regression, TyG index was positively associated with severe complications of COVID-19 (adjusted odds ratio: 1.42, 95% confidence interval [1.12–1.79]).
Conclusions
In COVID-19 infected patients, high TyG index was associated with increased risk for severe complications. TyG index might be useful predictor for the severity of COVID-19 infection.
Keywords: COVID-19, Insulin resistance, Triglyceride-glucose index, Glucose
Introduction
Coronavirus disease 2019 (COVID-19) is an infectious disease caused by the novel coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [1]. The global pandemic of COVID-19 is still ongoing and the most serious threat to public health in a century. Although most COVID-19 patients show good recovery, a number of patients, especially with underlying comorbidities, have critical complications including respiratory failure that can eventually lead to death [2]. Epidemiological studies have documented that the presence and number of comorbidities correlated with poorer outcome with COVID-19 infection [3]. Along with other comorbidities, diabetes mellitus was considered as a possible risk factor for COVID-19 infection and related complications [4].
Insulin resistance is the common metabolic disorder often associated with type 2 diabetes, representing a pathological condition in which the body's cells become resistant to the hormone insulin [5]. Insulin involves multiple molecular and pathophysiological pathways, and pathologic condition of insulin resistance is closely linked to not only diabetes mellitus, but also to various medical conditions such as hypertension, dyslipidemia, liver disease, cardiovascular disease, neurodegenerative disease, malignancy, obesity, inflammatory and infectious disease [6], [7]. In critically ill patients, insulin resistance has been reported to be correlated with the severity of condition and poor clinical outcome [8], [9], [10]. The triglyceride-glucose (TyG) index is a simple surrogate marker of insulin resistance, calculated using fasting triglyceride and fasting blood glucose [11]. High TyG index is significant indicator of adverse prognosis in general population as well as in patients with various diseases [12], [13]. We investigated whether TyG index measured before the COVID-19 infection has a prognostic value to identify those at high-risk for severe complications in COVID-19 infected patients.
Material and methods
Study design and participants
We performed a retrospective observational study for prognosis of COVID-19 infection according to TyG index with a nationwide population-based COVID-19 dataset in South Korea. South Korea has a single-payer public healthcare system, called by National Health Insurance Service (NHIS). NHIS has an information for hospital visit, diagnoses (recorded based on International Statistical Classification of Diseases and Related Health Problems 10th revision), medical procedure, prescription, and mortality of the whole Korean population [14]. Recently, the Korea Centers for Disease Control and Prevention, and the Ministry of Health and Welfare, South Korea released a nationwide COVID-19-related dataset, which includes the data of Korean citizens who underwent real-time reverse transcription polymerase chain reaction (RT-PCR) assays with nasal and pharyngeal swabs to COVID-19 from January 1 to June 4, 2020 [15]. The COVID-19-related dataset contains data on result of COVID-19 test (positive or negative), national health check-up program (2015–2018), health claims data, and mortality of all people who underwent RT-PCR tests for the COVID-19 [16]. The real-time PT PCR assays kit followed the WHO guideline and was validated by the Korea Centers for Disease Control and Prevention [17].
TyG index and covariates
As a public institution, the NHIS provides free health checkups to all Koreans aged ≥ 40 years every two years. The free health checkups program consisted of physical examination of blood pressure, body mass index (BMI), lifestyle survey (medical history, physical activity, smoking habit, alcohol consumption), and laboratory tests including fasting glucose, triglyceride, cholesterol levels [18]. The TyG index was calculated as ln [triglyceride (mg/dL) × fasting glucose level (mg/dL)/2], based on the blood chemistry data measured in the free nationwide health checkups program before the COVID-19 infection (2015–2018) [19]. In order to avoid errors due to extreme values, when the measured value ≥ the 99 percentile value or ≤ the 1 percentile value, they are replaced with the 99% percentile value or the 1 percentile value, subsequently. If study participants underwent twice or more health check-up during the period, the most recent data was used in this study.
From the health claims and health checkups program data, we also collected information for sex, age at COVID-19 diagnosis, household income level (four quartiles), alcohol consumption, smoking habit, physical activity, BMI, low density lipoprotein (LDL) cholesterol, and presence of comorbidities (hypertension, diabetes mellitus, stroke, coronary artery disease, chronic kidney disease, asthma, malignancy). BMI was calculated by one's weight (kg) divided by the square of height (m2). Hypertension was defined when participants received prescription blood pressure lowing agents (calcium-channel blocker, diuretics, angiotensin-converting enzyme inhibitor, angiotensin II receptor blocker, and beta-blockers) with hypertension diagnostic codes of ‘I10–15’, or had blood pressure ≥ 140/90 mmHg, or positive checking in a self-report questionnaire regarding hypertension [20]. Diabetes mellitus was defined when participants had received prescription anti-diabetic agents (sulphonylurea, metformin, meglitinide, thiazolidinedione, dipeptidyl peptidase-4 inhibitor, sodium-glucose co-transporter 2, glucagon-like peptide-1 agonists, α-glucosidase inhibitor, and insulin) with diabetes mellitus diagnostic codes of ‘E11–14’, or had fasting blood glucose> 7.0 mmol/L, or positive checking in a self-report questionnaire for diabetes mellitus. Stroke was defined as the presence of corresponding stroke diagnostic codes of ‘I60–I63’, ‘I69’. Coronary artery disease and chronic kidney disease were defined as the presence of corresponding diagnostic codes (coronary artery disease: I20–25, chronic kidney disease: N03, N05, N165, N18–19, N250, I12–13, Z490, Z491–2, Z940, Z992, E102, E112, E132, E142, or T861) [21], [22]. Asthma was determined by the presence of diagnostic codes of ‘J45–46’ as the primary diagnosis 2 times or more [23]. Malignancy was defined by having at least three times outpatient visits or once admission with the diagnostic code of ‘C00–97’ with specific cancer registration code of ‘V027’, ‘V193–4’. The data for smoking habit (current, former, never), alcohol consumption (average frequency per week:<1 time, 1–2 times, 3–4 times or ≥5 times), and physical activity (average frequency per week:<1 day, 1–4 days or ≥5 days) were obtained from the self-report questionnaire for life-styles in the health check-up program.
Study outcomes
We defined the primary outcome as a composite of 1) application of mechanical ventilation, 2) admission to the intensive care unit, 3) use of high-flow oxygen therapy, and 4) mortality within 2 months after COVID-19 diagnosis. Application of mechanical ventilation was identified by claim codes for mechanical ventilation (M5850, M5857, M5858, M5860). The admission to the intensive care unit was defined using the related claim codes for intensive care (AH110, AH150, AH180–5, AH190–5, AH210, AH250, AH280–9, AH28A, AH290–9, AH380–9, AH38A, AH390–9, AH501, AJ001–AJ011, AJ020–1, AJ031, AJ100–390, AJ2A0, AJ3A0, AJ500–590, V5100, V5200, V5210–20, V5500–5520). High-flow oxygen therapy was determined by the presence of claim code for high-flow nasal cannula therapy (M0046) [24]. Mortality data are obtained by NHIS, which have previously been validated [25].
Statistical analyses
For comparison of continuous and categorical variables between groups, independent t-tests and chi-square tests were performed, respectively. To evaluate the relationship between TyG index and severe complications of COVID-19 infection, we used univariable and multivariable logistic regression analyses. To detect and measure the collinearity in a multivariable regression model, we calculated variance inflation factor (VIF) for each variable. Small value of VIF indicates low correlation among variables and it is believed acceptable if it is less than 5 [26]. The subgroup analyses were conducted according to the sex, age (dichotomized more than 60 years old), history of hypertension, and history of diabetes mellitus. The results were presented as odds ratio (OR) and 95% confidence interval (CI). To help the understanding of the association between TyG index and COVID-19 outcomes, we draw a spline curve representing OR for primary outcome according to the continuous level of TyG index. All statistics were executed using R software, version 3.3.3 (R Foundation for Statistical Computing, Vienna, Austria), and SAS 9.4 version (SAS Inc., Cary, NC, USA). Two-sided P-values less than 0.05 were considered significant.
Results
Participants and demographics
The Korean nationwide COVID-19 cohort consisted of 212,678 participants aged ≥ 20 years who received RT-PCR test for COVID-19. Among them, 7713 patients who tested positive in the COVID-19 PCR test were included in this study, excluding those who tested negative in the COVID-19 PCR test (n = 204,965). Among these 7713 patients, there were 3934 patients who had an available data for national health screening program in 2015–2018. We additionally excluded 47 patients with incomplete data for analysis. Finally, 3887 patients with confirmed COVID-19 diagnosis were included in this study ( Fig. 1).
Fig. 1.
Flowchart depicting patient inclusion and exclusion. COVID-19, coronavirus disease 2019; RT-PCR, real-time reverse transcription polymerase chain reaction.
Among the finally included 3887 patients, 1451 (37.33%) were male and 1520 (39.10%) were ≥ 60 years of age. The mean± standard deviation of fasting glucose and triglyceride was 100.44 ± 22.87 mg/dL and 120.04 ± 73.46 mg/dL, respectively. Demographics, comorbidities, and laboratory findings are presented in Table 1. Compared to the group without severe complications, the severe complications of COVID-19 group were older and had a higher BMI (24.79 ± 3.43 vs. 23.91 ± 3.27 kg/m2, p < 0.001). Comorbidities such as hypertension, diabetes mellitus, stroke, and coronary artery disease, chronic kidney disease, asthma, and malignancy were more commonly noted in severe complications of COVID-19 group. The severe complications of COVID-19 group had a higher TyG index than the group without severe complications (8.81 ± 0.61 vs. 8.51 ± 0.61, p < 0.001), and fasting glucose level (111.64 ± 29.65 vs. 99.54 ± 22.00 mg/dL, p < 0.001) and triglyceride (142.58 ± 84.51 vs. 118.22 ± 72.20 mg/dL, p < 0.001) were higher (Table 1).
Table 1.
Baseline characteristics of COVID-19 patients according to prognosis.
| Variable | Total |
Primary outcomea |
P-valueb | |
|---|---|---|---|---|
| Without Severe complications of COVID-19 | With Severe complications of COVID-19 | |||
| Number of patients | 3887 | 3598 | 289 | |
| Sex, male | 1451 (37.33) | 1275 (35.44) | 176 (60.90) | < 0.001 |
| Age, years | < 0.001 | |||
| < 60 | 2367 (60.90) | 2307 (64.12) | 60 (20.76) | |
| ≥ 60 | 1520 (39.10) | 1291 (35.88) | 229 (79.24) | |
| Household income | 0.002 | |||
| Q1, lowest | 1320 (33.96) | 1241 (34.49) | 79 (27.34) | |
| Q2 | 912 (23.46) | 852 (23.68) | 60 (20.76) | |
| Q3 | 705 (18.14) | 651 (18.09) | 54 (18.69) | |
| Q4, highest | 950 (24.44) | 854 (23.74) | 96 (33.22) | |
| Alcohol consumption, frequency per week | < 0.001 | |||
| < 1 time | 2656 (68.33) | 2442 (67.87) | 214 (74.05) | |
| 1–2 times | 921 (23.69) | 880 (24.46) | 41 (14.19) | |
| 3–4 times | 225 (5.79) | 198 (5.50) | 27 (9.34) | |
| ≥ 5 times | 85 (2.19) | 78 (2.17) | 7 (2.42) | |
| Smoking habit | < 0.001 | |||
| None | 3018 (77.64) | 2840 (78.93) | 178 (61.59) | |
| Former smoker | 573 (14.74) | 481 (13.37) | 92 (31.83) | |
| Current smoker | 296 (7.62) | 277 (7.70) | 19 (6.57) | |
| Physical activity, frequency per week | 0.020 | |||
| < 1 day | 806 (20.74) | 730 (20.29) | 76 (26.30) | |
| 1–4 days | 1693 (43.56) | 1586 (44.08) | 107 (37.02) | |
| ≥ 5 days | 1388 (35.71) | 1282 (35.63) | 106 (36.68) | |
| Anthropometric measurements | ||||
| Body mass index (kg/m2) | 23.97 ± 3.29 | 23.91 ± 3.27 | 24.79 ± 3.43 | < 0.001 |
| Comorbidities | ||||
| Hypertension | 1304 (33.55) | 1115 (30.99) | 189 (65.40) | < 0.001 |
| Diabetes mellitus | 601 (15.46) | 502 (13.95) | 99 (34.26) | < 0.001 |
| Stroke | 156 (4.01) | 123 (3.42) | 33 (11.42) | < 0.001 |
| Coronary artery disease | 209 (5.38) | 172 (4.78) | 37 (12.80) | < 0.001 |
| Chronic kidney disease | 278 (7.15) | 230 (6.39) | 48 (16.61) | < 0.001 |
| Asthma | 212 (5.45) | 185 (5.14) | 29 (9.34) | 0.003 |
| Malignancy | 333 (8.57) | 289 (8.03) | 44 (15.22) | < 0.001 |
| Laboratory findings | ||||
| Fasting glucose, mg/dL | 100.44 ± 22.87 | 99.54 ± 22.00 | 111.64 ± 29.65 | < 0.001 |
| Triglyceride, mg/dL | 120.04 ± 73.46 | 118.22 ± 72.20 | 142.58 ± 84.51 | < 0.001 |
| LDL cholesterol, mg/dL | 114.23 ± 33.03 | 114.83 ± 32.84 | 106.74 ± 34.46 | < 0.001 |
| TyG index | 8.54 ± 0.61 | 8.51 ± 0.61 | 8.81 ± 0.61 | < 0.001 |
| Clinical outcome | ||||
| Mechanical ventilation | 119 (3.06) | |||
| Intensive care unit care | 171 (4.40) | |||
| High-flow oxygen therapy | 83 (2.14) | |||
| Mortality | 114 (2.93) | |||
Data are represented as number of patients (%) or mean± standard deviation. Q, quartile; HDL, High-density lipoprotein; LDL, Low-density lipoprotein; TyG index is calculated by the following formula: ln [fasting triglycerides (mg/dL) × fasting glucose (mg/dL)/2].
Primary outcome is composite of mechanical ventilation, intensive care unit care, high-flow oxygen therapy, and mortality within two months after COVID-19 diagnosis
P-value is derived from independent t-test or Chi’s square test between patient groups according to primary outcome.
Association of TyG index with severe complications of COVID-19
Primary outcome for severe complications of COVID-19 was noted in 289 (7.44%) patients including mechanical ventilation in 119 (3.06%) patients, intensive care unit care in 171 (4.40%) patients, high-flow oxygen therapy in 83 (2.14%) patients, and mortality in 114 (2.93%) patients, respectively. After adjusting sex, age, household income, alcohol consumption, smoking habit, physical activity, BMI, hypertension, diabetes mellitus, stroke, coronary artery disease, chronic kidney disease, asthma, malignancy, and LDL cholesterol, TyG index was positively associated with severe complications of COVID-19 (OR: 1.42, 95% CI [1.12 – 1.79], p = 0.003) ( Table 2). In a testing for multicollinearity with VIF, all values are< 2 indicating absence of high multicollinearity.
Table 2.
Association factors for occurrence of severe complications in COVID-19 patients.
| Variable | Univariate OR [95% CI] | P-value | Adjusted OR [95% CI]a | P-value |
|---|---|---|---|---|
| Sex, male | 2.84 [2.22–3.63] | < 0.001 | 2.40 [1.72–3.34] | < 0.001 |
| Age, years | ||||
| < 60 | 1 (Ref) | – | 1 (Ref) | – |
| ≥ 60 | 6.82 [5.09–9.14] | < 0.001 | 3.89 [2.81–5.37] | < 0.001 |
| Household income | ||||
| Q1, lowest | 1 (Ref) | – | 1 (Ref) | – |
| Q2 | 1.11 [0.78–1.57] | 0.568 | 1.15 [0.79–1.66] | 0.474 |
| Q3 | 1.30 [0.91–1.87] | 0.148 | 1.17 [0.80–1.72] | 0.422 |
| Q4, highest | 1.77 [1.30–2.41] | < 0.001 | 1.42 [1.02–1.99] | 0.040 |
| Alcohol consumption, frequency per week | ||||
| < 1 time | 1 (Ref) | – | 1 (Ref) | – |
| 1–2 times | 0.53 [0.38–0.75] | < 0.001 | 0.52 [0.35–0.77] | 0.001 |
| 3–4 times | 1.56 [1.02–2.38] | 0.042 | 0.95 [0.58–1.57] | 0.851 |
| ≥ 5 times | 1.02 [0.47–2.25] | 0.953 | 0.46 [0.20–1.07] | 0.072 |
| Smoking habit | ||||
| None | 1 (Ref) | – | 1 (Ref) | – |
| Former smoker | 3.05 [2.33–4.00] | < 0.001 | 1.53 [1.07–2.21] | 0.021 |
| Current smoker | 1.09 [0.67–1.78] | 0.718 | 0.67 [0.38–1.17] | 0.155 |
| Physical activity, frequency per week | ||||
| < 1 day | 1 (Ref) | – | 1 (Ref) | – |
| 1–4 days | 0.65 [0.48–0.88] | 0.006 | 0.77 [0.55–1.07] | 0.124 |
| ≥ 5 days | 0.79 [0.58–1.08] | 0.142 | 0.80 [0.57–1.12] | 0.199 |
| Anthropometric measurements | ||||
| Body mass index (kg/m2) | 1.08 [1.04–1.12] | < 0.001 | 1.01 [0.97–1.06] | 0.586 |
| Comorbidities | ||||
| Hypertension | 4.21 [3.27–5.42] | < 0.001 | 1.78 [1.33–2.39] | < 0.001 |
| Diabetes mellitus | 3.21 [2.48–4.17] | < 0.001 | 1.20 [0.87–1.66] | 0.261 |
| Stroke | 3.64 [2.43–5.46] | < 0.001 | 1.25 [0.80–1.96] | 0.326 |
| Coronary artery disease | 2.93 [2.01–4.27] | < 0.001 | 1.21 [0.79–1.85] | 0.386 |
| Chronic kidney disease | 2.92 [2.08–4.09] | < 0.001 | 1.27 [0.86–1.88] | 0.225 |
| Asthma | 1.90 [1.25–2.90] | 0.003 | 1.33 [0.84–2.12] | 0.222 |
| Malignancy | 2.06 [1.46–2.90] | < 0.001 | 1.36 [0.93–1.99] | 0.115 |
| Laboratory findings | ||||
| LDL cholesterol, mmol/L | 0.87 [0.82–0.93] | < 0.001 | 0.93 [0.86–0.99] | 0.032 |
| Triglyceride/glucose index | 2.11 [1.75–2.54] | < 0.001 | 1.42 [1.12–1.79] | 0.003 |
Data are derived from logistic regression analysis for severe complications of COVID-19. Severe complications of COVID-19 are composite of mechanical ventilation, intensive care unit care, high-flow oxygen therapy, and mortality.
OR, odds ratio; CI, confidence interval; Q, quartile; LDL, low-density lipoprotein.
Adjusted for sex, age, household income, alcohol consumption, smoking status, physical activity, body mass index, hypertension, diabetes mellitus, stroke, coronary artery disease, chronic kidney disease, asthma, malignancy, and LDL cholesterol.
Distribution of TyG index according to the presence of severe complications of COVID-19 are demonstrated in Fig. 2. A spline curve demonstrated the positive association between TyG index and severe complications of COVID-19 ( Fig. 3). In subgroup analysis by sex, age, hypertension, and diabetes mellitus, the positive association of TyG index with severe complications of COVID-19 was consistently noted (all p for interaction>0.05, Fig. 4).
Fig. 2.
Distribution of triglyceride-glucose index according to presence of severe complications of COVID-19. Vertical dotted lines present the mean values of triglyceride-glucose index in each groups with and without severe complications of COVID-19.
Fig. 3.
Spline curve of the association between triglyceride-glucose index and severe complications of COVID-19. The black lines and gray shadows represent the odds ratio and 95% confidence intervals for the presence of severe complications of COVID-19 (composite of mechanical ventilation, intensive care unit care, high-flow oxygen therapy, and mortality). Ref indicates median value (8.49) of triglyceride-glucose index. TyG, triglyceride-glucose index; CI: confidence interval.
Fig. 4.
Triglyceride-glucose index on the severe complications of COVID-19 in each subgroup by risk factor. Data include odds ratio (OR) and 95% confidence intervals (CI) for triglyceride-glucose index derived from logistic regression model for presence of severe complications in COVID-19 patients. P-values for interaction are calculated between each risk factor and triglyceride-glucose index.
Discussion
The key finding of our study based on a nationwide population-based COVID-database in South Korea is that high TyG index, which represents insulin resistance, was associated with the increased risk for severe complications of COVID-19 patients.
In a previous hospital-based study with 151 COVID-19 positive patients, the TyG index levels were significantly higher in the patients with severe complication or mortality [27]. Among patients hospitalized for COVID-19, those with critical COVID-19 presented higher levels of fasting glucose, triglycerides, both of which are components of the TyG index [28]. Another hospital-based case-control study with 43 patients revealed that biochemical markers of insulin resistance may be related with severe complications of COVID-19 [29].
Our result is in lines with these previous studies for the potential role of TyG index and markers of insulin resistance as an indicator for the severe complications of COVID-19 [27]. Unlike the prior studies, we calculated TyG index based on a nationwide health screening data measured before the COVID-19 infection. It has been known that COVID-19 infection itself can induce insulin resistance and lipid metabolic dysregulation [30], [31]. The previous case series demonstrated that COVID-19 infection triggered uncontrolled hyperglycemia and insulin resistance in acute phase [32], [33], [34]. Therefore, worse prognosis of COVID-19 patients with high TyG index measured in acute phase might be due to that critical illness with COVID-19 infection itself cause further insulin resistance [35]. We found that the TyG index measured before the COVID-19 infection also had a significant association with COVID-19 prognosis. This result added evidence that underlying insulin resistance in health condition is one of determinant for the prognosis of COVID-19 infection. Regarding the interplay between inflammation and insulin resistance, baseline insulin resistance may hold key to improving immune responses against COVID-19 [36]. In addition, our study showed that TyG index easily calculated from a pre-existing blood test can serve as a simple tool to predict the risk of severe complications from COVID-19 without additional test. It will be applicable to screening those at high-risk in general population who need to pay more attention to COVID-19 infection and determining priority group for COVID-19 vaccination.
Although our study is not a study for the mechanism, we may allow proposing hypotheses that may explain the association between the TyG index and severe COVID-19 outcome. TyG index was associated with metabolic syndrome, dyslipidemia, future development of diabetes mellitus, subclinical coronary artery disease, arterial stiffness, and occurrence of major cardiovascular events [11], [37], [38]. Epidemiologic data have documented that individuals with high TyG index are at increased risk for nonalcoholic fatty liver disease, which is also closely related to obesity and insulin resistance [39], [40]. It has been established that the presence and number of these comorbidities in COVID-19 patients, which closely related with insulin resistance and TyG index, are associated with poor prognosis of COVID-19 [38], [41], [42]. Because TyG index is considered a good surrogate marker of the comorbidities, those with high TyG index may have more risk factors and worse clinical outcomes in COVID-19 patients [27].
In COVID-19 infection, immunologic response plays a key role to maintain health and pathogenesis [43]. Current evidence suggested that dysregulated immune response may be responsible for COVID-19 pathogenesis and related critical illness, including lymphopenia, neutrophilia, dysregulation of monocytes/macrophages, impaired type I interferon response, antibody-dependent enhancement, and cytokine storm [44], [45]. Studies in insulin resistance have revealed a clear association between the chronic activation of pro-inflammatory signaling pathways/impaired immune response and decreased insulin sensitivity [46], [47]. Macrophages chronically exposed to high levels of insulin result in sustained inflammatory cytokine production and altered response to pathogen [48]. It has become apparent that the impaired immune response through multiple pathways in insulin resistance can lead to systemic alterations contributing to excessive production of pro-inflammatory cytokines, also known as a cytokine storm [36], [49]. Because hyper-activated inflammatory responses with cytokine storm occurring in COVID-19 are regarded major pathogenesis of critical illness with COVID-19 [50], chronic activation of pro-inflammatory signaling and altered immunologic response with insulin resistance, presented as high TG index, could be a predisposing factor for severe complications of COVID-19 [27]. Accordingly, viremia may be worsened by insulin resistance, and this increased risk of viremia may cause poor prognosis or morality after COVID-19 infection [51], [52].
Our study has limitations. First, because our study had a retrospective cohort design, the causal relationship could not be proven. Second, in the health screening cohort, young people under the age of 40 could not be included, so there would be a selection bias. Third, it is difficult to generalize our results for overall ethnicity because our dataset consists of only the Korean general population. Fourth, because our dataset checked the lipid profile before the COVID-19 infection, there was a lack of the serial laboratory finding or the change of the lipid profile after COVID-19. Fifth, concurrent medications before and after COVID-19 infection might be a possible confounding factor on the prognosis, which were not evaluated in current study. Since the TyG index is the calculated product of fasting triglyceride level and fasting blood glucose level, it could be affected by the anti-diabetic or lipid-lowering agents taken at the time of health screening. Finally, since insulin levels were not measured in our nationwide screening dataset, we could not directly check the insulin resistance.
In conclusion, our study demonstrated that TyG index measured before the COVID-19 infection was significantly associated with increased risk for severe complications of COVID-19. TyG index, which is simply calculated from fasting triglyceride and fasting blood glucose, may be a useful prognostic biomarker in COVID-19 infection. Further study is need for the pathophysiologic linking between insulin resistance and COVID-19 adverse outcomes.
Funding
This study was supported by a new faculty research seed money grant of Yonsei University College of Medicine for 2022 (2022-32-0058). This work was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2020R1I1A1A01060447 to JK, 2021R1F1A1048113 to T-JS, 2021R1I1A1A01059868 to YC). The funders had no role in study design, data collection and analysis, the decision to publish, or preparation of the manuscript.
Ethics approval
The NHIS data was fully anonymized for privacy protection. Due to the retrospective analysis based on the fully anonymized data from our dataset, this study was approved by the Institutional Review Board of our institution (Seoul Hospital Ewha Womans University College of Medicine 2020-10-021), and the requirement for informed consent was waived.
Declaration of Competing Interest
The authors declare no conflicts of interest.
Acknowledgments
This study used the COVID-19 dataset (NHIS-2021-1-347) created by NHIS.
Author’s contributions
Tae-Jin Song and Yoonkyung Chang designed the research and drafted the manuscript. Jinkwon Kim and Jimin Jeon performed the statistical analysis and the interpretation of data. Tae-Jin Song and Jinkwon Kim participated in the interpretation of all data and the critical revision of the manuscript for intellectual content. All authors read and approved the manuscript.
Data availability
The COVID-19 database used in the present study is available from the National Health Insurance Sharing Service [http://nhiss.nhis.or.kr/bd/ab/bdaba021eng.do], but restrictions apply to the availability of these data, which were used under licence for the current study and so are not publicly available. To access the data, a complete application form, research proposal, and an approval document for research purposes from an institutional review board should be submitted to the NHIS for review by the inquiry committee of research support.
References
- 1.Hu B., Guo H., Zhou P., Shi Z.L. Characteristics of SARS-CoV-2 and COVID-19. Nat Rev Microbiol. 2021;19(3):141–154. doi: 10.1038/s41579-020-00459-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.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(2):e16–e25. doi: 10.1016/j.jinf.2020.04.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Liu W., Yang C., Liao Y.G., Wan F., Lin L., Huang X., et al. Risk factors for COVID-19 progression and mortality in hospitalized patients without pre-existing comorbidities. J Infect Public Health. 2022;15(1):13–20. doi: 10.1016/j.jiph.2021.11.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lim S., Bae J.H., Kwon H.S., Nauck M.A. COVID-19 and diabetes mellitus: from pathophysiology to clinical management. Nat Rev Endocrinol. 2021;17(1):11–30. doi: 10.1038/s41574-020-00435-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lebovitz H.E. Insulin resistance: definition and consequences. Exp Clin Endocrinol Diabetes. 2001;109(Suppl 2):S135–S148. doi: 10.1055/s-2001-18576. [DOI] [PubMed] [Google Scholar]
- 6.Singh B., Saxena A. Surrogate markers of insulin resistance: a review. World J Diabetes. 2010;1(2):36–47. doi: 10.4239/wjd.v1.i2.36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bugianesi E., McCullough A.J., Marchesini G. Insulin resistance: a metabolic pathway to chronic liver disease. Hepatology. 2005;42(5):987–1000. doi: 10.1002/hep.20920. [DOI] [PubMed] [Google Scholar]
- 8.Ilias I., Diamantopoulos A., Pratikaki M., Botoula E., Jahaj E., Athanasiou N., et al. Glycemia, beta-cell function and sensitivity to insulin in mildly to critically Ill covid-19 patients. Medicina. 2021;57(1) doi: 10.3390/medicina57010068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Trifunovic D., Stankovic S., Sobic-Saranovic D., Marinkovic J., Petrovic M., Orlic D., et al. Acute insulin resistance in ST-segment elevation myocardial infarction in non-diabetic patients is associated with incomplete myocardial reperfusion and impaired coronary microcirculatory function. Cardiovasc Diabetol. 2014;13:73. doi: 10.1186/1475-2840-13-73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Preiser J.C., Ichai C., Orban J.C., Groeneveld A.B. Metabolic response to the stress of critical illness. Br J Anaesth. 2014;113(6):945–954. doi: 10.1093/bja/aeu187. [DOI] [PubMed] [Google Scholar]
- 11.Lee S.B., Ahn C.W., Lee B.K., Kang S., Nam J.S., You J.H., et al. Association between triglyceride glucose index and arterial stiffness in Korean adults. Cardiovasc Diabetol. 2018;17(1):1–6. doi: 10.1186/s12933-018-0692-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Park K., Ahn C.W., Lee S.B., Kang S., Nam J.S., Lee B.K., et al. Elevated TyG index predicts progression of coronary artery calcification. Diabetes Care. 2019;42(8):1569–1573. doi: 10.2337/dc18-1920. [DOI] [PubMed] [Google Scholar]
- 13.Liu X.C., He G.D., Lo K., Huang Y.Q., Feng Y.Q. The triglyceride-glucose index, an insulin resistance marker, was non-linear associated with all-cause and cardiovascular mortality in the general population. Front Cardiovasc Med. 2020;7 doi: 10.3389/fcvm.2020.628109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Seong S.C., Kim Y.Y., Park S.K., Khang Y.H., Kim H.C., Park J.H., et al. Cohort profile: the National Health Insurance Service-National Health Screening Cohort (NHIS-HEALS) in Korea. BMJ Open. 2017;7(9) doi: 10.1136/bmjopen-2017-016640. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kim S.H., Choi Y., Lee D., Lee H., Kim J.H., Choi E.S., et al. Impact of COVID-19 on pregnant women in South Korea: focusing on prevalence, severity, and clinical outcomes. J Infect Public Health. 2022;15(2):270–276. doi: 10.1016/j.jiph.2022.01.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lee S.W., Yang J.M., Moon S.Y., Yoo I.K., Ha E.K., Kim S.Y., et al. Association between mental illness and COVID-19 susceptibility and clinical outcomes in South Korea: a nationwide cohort study. Lancet Psychiatry. 2020;7(12):1025–1031. doi: 10.1016/s2215-0366(20)30421-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lee S.W., Ha E.K., Yeniova A., Moon S.Y., Kim S.Y., Koh H.Y., et al. Severe clinical outcomes of COVID-19 associated with proton pump inhibitors: a nationwide cohort study with propensity score matching. Gut. 2021;70(1):76–84. doi: 10.1136/gutjnl-2020-322248. [DOI] [PubMed] [Google Scholar]
- 18.Chang Y., Lee J.S., Lee K.J., Woo H.G., Song T.J. Improved oral hygiene is associated with decreased risk of new-onset diabetes: a nationwide population-based cohort study. Diabetologia. 2020;63(5):924–933. doi: 10.1007/s00125-020-05112-9. [DOI] [PubMed] [Google Scholar]
- 19.Hong S., Han K., Park C.Y. The triglyceride glucose index is a simple and low-cost marker associated with atherosclerotic cardiovascular disease: a population-based study. BMC Med. 2020;18(1):361. doi: 10.1186/s12916-020-01824-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kim J., Bushnell C.D., Lee H.S., Han S.W. Effect of adherence to antihypertensive medication on the long-term outcome after hemorrhagic stroke in Korea. Hypertension. 2018;72(2):391–398. doi: 10.1161/HYPERTENSIONAHA.118.11139. [DOI] [PubMed] [Google Scholar]
- 21.Jung C.H., Seo G.H., Suh S., Bae J.C., Kim M.K., Hwang Y.C., et al. The population-based risk of need for coronary revascularization according to the presence of type 2 diabetes mellitus and history of coronary heart disease in the Korean population. PLoS One. 2015;10(6) doi: 10.1371/journal.pone.0128627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kang S.H., Kim S.W., Kim A.Y., Cho K.H., Park J.W., Do J.Y. Association between chronic kidney disease or acute kidney injury and clinical outcomes in COVID-19 patients. J Korean Med Sci. 2020;35(50) doi: 10.3346/jkms.2020.35.e434. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kim J.H., Wee J.H., Choi H.G., Park J.Y., Hwang Y.I., Jang S.H., et al. Association between statin medication and asthma/asthma exacerbation in a national health screening cohort. J Allergy Clin Immunol Pract. 2021;9(7):2783–2791. doi: 10.1016/j.jaip.2021.04.014. [DOI] [PubMed] [Google Scholar]
- 24.Son M., Seo J., Yang S. Association between renin-angiotensin-aldosterone system inhibitors and COVID-19 infection in South Korea. Hypertension. 2020;76(3):742–749. doi: 10.1161/HYPERTENSIONAHA.120.15464. [DOI] [PubMed] [Google Scholar]
- 25.Song T.J., Jeon J., Kim J. Cardiovascular risks of periodontitis and oral hygiene indicators in patients with diabetes mellitus. Diabetes Metab. 2021;47(6) doi: 10.1016/j.diabet.2021.101252. [DOI] [PubMed] [Google Scholar]
- 26.Neter J., Kutner M.H., Nachtsheim C.J., Wasserman W. Applied linear statistical models. 1996.
- 27.Ren H., Yang Y., Wang F., Yan Y., Shi X., Dong K., et al. Association of the insulin resistance marker TyG index with the severity and mortality of COVID-19. Cardiovasc Diabetol. 2020;19(1):58. doi: 10.1186/s12933-020-01035-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Bellia A., Andreadi A., Giudice L., De Taddeo S., Maiorino A., D’Ippolito I., et al. Atherogenic dyslipidemia on admission is associated with poorer outcome in people with and without diabetes hospitalized for COVID-19. Diabetes Care. 2021;44(9):2149–2157. doi: 10.2337/dc20-2838. [DOI] [PubMed] [Google Scholar]
- 29.Alcántara-Alonso E., Molinar-Ramos F., González-López J.A., Alcántara-Alonso V., Muñoz-Pérez M.A., Lozano-Nuevo J.J., et al. High triglyceride to HDL-cholesterol ratio as a biochemical marker of severe outcomes in COVID-19 patients. Clin Nutr Espen. 2021:437–444. doi: 10.1016/j.clnesp.2021.04.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Govender N., Khaliq O.P., Moodley J., Naicker T. Insulin resistance in COVID-19 and diabetes. Prim Care Diabetes. 2021;15(4):629–634. doi: 10.1016/j.pcd.2021.04.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Chen M., Zhu B., Chen D., Hu X., Xu X., Shen W.J., et al. COVID-19 may increase the risk of insulin resistance in adult patients without diabetes: a 6-month prospective study. Endocr Pract. 2021;27(8):834–841. doi: 10.1016/j.eprac.2021.04.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Affinati A.H., Wallia A., Gianchandani R.Y. Severe hyperglycemia and insulin resistance in patients with SARS-CoV-2 infection: a report of two cases. Clin Diabetes Endocrinol. 2021;7(1):8. doi: 10.1186/s40842-021-00121-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Seggelke S.A., Ingram C.C., Crawley S., Low Wang C.C. Insulin resistance in a hospitalized COVID-19 patient: a case review. Clin Diabetes. 2021;39(2):228–232. doi: 10.2337/cd20-0036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Jornayvaz F.R., Assouline B., Pugin J., Gariani K. Extremely high-dose insulin requirement in a diabetic patient with COVID-19: a case report. BMC Endocr Disord. 2020;20(1):155. doi: 10.1186/s12902-020-00632-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Langouche L., Van den Berghe G., Gunst J. Hyperglycemia and insulin resistance in COVID-19 versus non-COVID critical illness: are they really different. Crit Care. 2021;25(1):437. doi: 10.1186/s13054-021-03861-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Gangadharan C., Ahluwalia R., Sigamani A. Diabetes and COVID-19: role of insulin resistance as a risk factor for COVID-19 severity. World J Diabetes. 2021;12(9):1550–1562. doi: 10.4239/wjd.v12.i9.1550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Chamroonkiadtikun P., Ananchaisarp T., Wanichanon W. The triglyceride-glucose index, a predictor of type 2 diabetes development: a retrospective cohort study. Prim Care Diabetes. 2020;14(2):161–167. doi: 10.1016/j.pcd.2019.08.004. [DOI] [PubMed] [Google Scholar]
- 38.Park G.M., Cho Y.R., Won K.B., Yang Y.J., Park S., Ann S.H., et al. Triglyceride glucose index is a useful marker for predicting subclinical coronary artery disease in the absence of traditional risk factors. Lipids Health Dis. 2020;19(1):7. doi: 10.1186/s12944-020-1187-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Kitae A., Hashimoto Y., Hamaguchi M., Obora A., Kojima T., Fukui M. The Triglyceride and glucose index is a predictor of incident nonalcoholic fatty liver disease: a population-based cohort study. Can J Gastroenterol Hepatol. 2019;2019 doi: 10.1155/2019/5121574. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Zhang S., Du T., Zhang J., Lu H., Lin X., Xie J., et al. The triglyceride and glucose index (TyG) is an effective biomarker to identify nonalcoholic fatty liver disease. Lipids Health Dis. 2017;16(1):15. doi: 10.1186/s12944-017-0409-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Luo E., Wang D., Yan G., Qiao Y., Liu B., Hou J., et al. High triglyceride-glucose index is associated with poor prognosis in patients with acute ST-elevation myocardial infarction after percutaneous coronary intervention. Cardiovasc Diabetol. 2019;18(1):150. doi: 10.1186/s12933-019-0957-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Liu D., Zheng Y., Kang J., Wang D., Bai L., Mao Y., et al. Not only high number and specific comorbidities but also age are closely related to progression and poor prognosis in patients with COVID-19. Front Med. 2021;8 doi: 10.3389/fmed.2021.736109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Chowdhury M.A., Hossain N., Kashem M.A., Shahid M.A., Alam A. Immune response in COVID-19: a review. J Infect Public Health. 2020;13(11):1619–1629. doi: 10.1016/j.jiph.2020.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Yang L., Xie X., Tu Z., Fu J., Xu D., Zhou Y. The signal pathways and treatment of cytokine storm in COVID-19. Signal Transduct Target Ther. 2021;6(1):255. doi: 10.1038/s41392-021-00679-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Sharma S.K., Bhatt P., Asdaq S.M.B., Alshammari M.K., Alanazi A., Alrasheedi N.S., et al. Combined therapy with ivermectin and doxycycline can effectively alleviate the cytokine storm of COVID-19 infection amid vaccination drive: a narrative review. J Infect Public Health. 2022 doi: 10.1016/j.jiph.2022.03.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.de Luca C., Olefsky J.M. Inflammation and insulin resistance. FEBS Lett. 2008;582(1):97–105. doi: 10.1016/j.febslet.2007.11.057. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Rehman K., Akash M.S. Mechanisms of inflammatory responses and development of insulin resistance: how are they interlinked. J Biomed Sci. 2016;23(1):87. doi: 10.1186/s12929-016-0303-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Ieronymaki E., Daskalaki M.G., Lyroni K., Tsatsanis C. Insulin signaling and insulin resistance facilitate trained immunity in macrophages through metabolic and epigenetic changes. Front Immunol. 2019;10:1330. doi: 10.3389/fimmu.2019.01330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Damiati L.A., Bahlas S., Aljohaney A., Bawazir Y., Mustafa M., Denetiu I., et al. Implications of SARS-CoV-2 infection on the clinical, hematological, and inflammatory parameters in COVID-19 patients: a retrospective cross-sectional study. J Infect Public Health. 2022;15(2):214–221. doi: 10.1016/j.jiph.2021.12.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Merad M., Martin J.C. Pathological inflammation in patients with COVID-19: a key role for monocytes and macrophages. Nat Rev Immunol. 2020;20(6):355–362. doi: 10.1038/s41577-020-0331-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Finucane F.M., Davenport C. Coronavirus and obesity: could insulin resistance mediate the severity of Covid-19 infection. Front Public Health. 2020;8(184) doi: 10.3389/fpubh.2020.00184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Myhre P.L., Prebensen C., Jonassen C.M., Berdal J.E., Omland T. SARS-CoV-2 viremia is associated with inflammatory, but not cardiovascular biomarkers, in patients hospitalized for COVID-19. J Am Heart Assoc. 2021;10(9) doi: 10.1161/JAHA.120.019756. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The COVID-19 database used in the present study is available from the National Health Insurance Sharing Service [http://nhiss.nhis.or.kr/bd/ab/bdaba021eng.do], but restrictions apply to the availability of these data, which were used under licence for the current study and so are not publicly available. To access the data, a complete application form, research proposal, and an approval document for research purposes from an institutional review board should be submitted to the NHIS for review by the inquiry committee of research support.