Prognostic Value of Hyperglycemia and Insulin Resistance Among Patients with Confirmed COVID-19 Infections at Admission to the Alexandria Fever Hospital,Egypt

Abstract

Background:

The interaction between COVID-19 infection, hyperglycemia, and insulin resistance (IR) may lead to poor outcome.

Methods:

This prospective study included 100 adult participants without diabetes attending Alexandria Fever Hospital with confirmed COVID-19 infections. They were classified into four groups according to disease severity using World Health Organization (WHO) criteria. Demographic and clinical parameters were collected. Laboratory investigations were obtained. Another follow-up fasting plasma glucose (FPG) value was measured after 3 months in cured patients.

Results:

Admission FPG, follow-up FPG, lipid profile, markers of IR, and inflammation were significantly higher in severe and critical cases than in mild and moderate cases with increasing values with increased severity. Furthermore, these parameters were significantly higher in died cases compared with cured cases. Admission FPG, TyG index, and homeostatic model assessment (HOMA)-IR showed significant positive correlations with follow-up FPG. Admission FPG was the only independent mortality predictor in multivariate analysis (P = 0.027) with 1.7-folds increased mortality risk with each 10 mg/dL increments. Values exceeding 117 mg/dL, 2.2, and 6.33 for admission FPG, HOMA-IR, and Fasting Insulin Resistance Index, respectively, were able to predict mortality in the studied sample.

Conclusions:

These results will help in identifying patients at high risk of severe infection and death at admission and take early actions to improve outcome.

Introduction

The recent global pandemic of coronavirus disease 19 (COVID-19) was associated with several co-morbidities. Its relation to diabetes mellitus (DM) and hyperglycemia includes many aspects. Previous studies revealed a bidirectional relationship between DM and COVID-19 infection. Patients with DM have worse prognosis of COVID-19 infection especially if uncontrolled. In contrast, COVID-19 infection leads to poor DM control. However, hyperglycemia per se without pre-existing DM was associated with poor outcome in patients infected with COVID-19.^1,2

New onset hyperglycemia that occurs at hospital admission in COVID-19 patients could be classified into stress-induced hyperglycemia, previously unrecognized new onset DM, direct effect of SARS-CoV-2 on endocrine system, and corticosteroids-induced hyperglycemia.^3,4

The pathogenesis of COVID-19 infection-associated hyperglycemia is multifactorial. First, hyperglycemia leads to increased angiotensin converting enzyme receptors on epithelial cells, which enhances interstitial pneumonitis and acute respiratory distress syndrome (ARDS).⁵ Second, hyperglycemia induces immune system over activation and abnormal inflammatory response.⁶ Furthermore, hyperglycemia induces elevated lactate level that leads to inflammatory immune system regulation and also hyperglycemia is associated with elevated lactate dehydrogenase level.⁷

COVID-19 infection is characterized by profound inflammatory state with elevated levels of C-reactive protein (CRP), D-dimer, ferritin, and interleukin 6 (IL-6). The exaggerated form of the immune response may lead to cytokine storm that induces widespread tissue damage and multi-organ failure. In contrast, DM and hyperglycemia are associated with a state of low-grade inflammation with elevation of inflammatory markers.^8,9

Insulin resistance (IR) is another important metabolic condition associated with hyperglycemia and obesity. It was found that during COVID-19 infection, angiotensin II is increased leading to oxidative stress, IR and cardiac dysfunction. The co-existence of hyperglycemia, IR, and COVID-19 infection will lead to a state of hyperinflammation, which in turn is associated with morbidity and mortality.^10,11

The aim of this study was to evaluate the prognostic value of hyperglycemia and IR among patients with confirmed COVID-19 infections at admission.

Materials and Methods

Subjects

This prospective study included 100 adult participants without DM attending the emergency department of Alexandria Fever Hospital, Alexandria, Egypt with confirmed COVID-19 infections (with positive real-time reverse transcription polymerase chain reaction nasopharyngeal sample result for SARS-CoV-2). All participants completed their hospital course with either discharge or death. Participants were divided into four equal groups of 25 patients, including mild, moderate, severe, and critical groups according to the severity of COVID-19 infection.¹² Exclusion criteria included pre-existing DM, pregnant women, age <20 years, end-stage renal disease, advanced liver disease, or received corticosteroids before admission.

Definitions and variables

Demographic data, including age and gender, were obtained from participants. Routine investigations were done, including complete blood count, liver function tests, renal functions tests, and lipid profile. The diagnosis, classification, and discharge criteria for COVID-19 patients were based on the guidelines issued by the World Health Organization (WHO).¹² The severity was classified as follows: (1) mild: mild symptoms, no pneumonia in imaging diagnosis; (2) moderate: fever, respiratory tract symptoms, and pneumonia in imaging diagnosis; (3) severe: either respiratory rate ≥30 breath/min, or finger oxygen saturation ≤93% at rest, or arterial blood oxygen partial pressure/oxygen concentration ≤300 mmHg; or (4) critical: respiratory failure requiring mechanical ventilation, organ failure requiring care in the intensive care unit, or shock.

The following parameters were measured: glucose-specific parameters¹³: fasting plasma glucose (FPG) level on admission (mg/dL), serum insulin level, HbA1c. Another follow-up FPG test was obtained in follow-up after 2–3 months. Regarding IR parameters; homeostatic model assessment-insulin resistance (HOMA-IR),¹⁴ Fasting Insulin Resistance Index (FIRI),¹⁵ Quantitative Insulin Sensitivity Check Index (QUICKI),¹⁵ and triglyceride-glucose (TyG) index¹⁶ were calculated.

In addition, inflammatory markers, including serum high sensitivity C-reactive protein (hs-CRP), IL 6, serum ferritin, and D-dimer were measured.^17,18

Ethical approval

The study was approved by the ethical committee of Faculty of Medicine, Alexandria University. It was conducted according to ethical standards stated in Helsinki Declaration. Written informed consent was obtained from each participant after explaining the nature of the study.

Statistical analysis

IBM SPSS software package version 20.0 (IBM Corp, Armonk, NY, USA) was used to analyze data. Number and percent were used to describe qualitative data. The normality of distribution was verified using Shapiro–Wilk test. Mean and standard deviation were used to describe quantitative data. Significance of the obtained results was judged at the 5% level. Chi-squared test was used to compare between different groups for categorical variables. F-test (analysis of variance) to compare between more than two groups, and post hoc test (Tukey) for pairwise comparisons for normally distributed quantitative variables. Kruskal–Wallis test was used to compare between more than two studied groups for abnormally distributed quantitative variables, whereas post hoc (Dunn's multiple comparisons test) was used for pairwise comparisons.

Spearman coefficient was used to correlate between two abnormally distributed quantitative variables. Regression analysis was performed to detect predictors of mortality in the studied population. Receiver operating characteristic (ROC) curve was generated by plotting sensitivity (TP) on Y-axis versus 1-specificity (FP) on X-axis at different cutoff values. The area under the ROC curve (AUC) denotes the diagnostic performance of the test. Acceptable performance is defined as area more than 50% gives and the best performance of the test is achieved at area about 100%.

Results

This study included 100 patients with confirmed COVID-19 infection admitted to Alexandria Fever Hospital. The total sample was divided into four equal groups according to severity. Comparison between the four studied groups is shown in Table 1. Regarding outcome, all patients in mild and moderate groups had complete cure. However, 12 (48%) patients in severe group and only 5 (20%) patients in critical group were cured, and the rest of the patients died. Patients remained in the same group till the end of the follow-up period.

Table 1.

Comparison Between the Four Studied Groups Regarding Different Variables

Demographic data, Mean ± SD	Mild (n = 25)	Moderate (n = 25)	Severe (n = 25)	Critical (n = 25)	P
Male No. (%)	12 (48)	14 (56)	13 (52)	13 (52)	0.956
Age (years) median (IQR)	42.0(33.0–55.0)	47.0(35.0–55.0)	59.0(44.0–65.0)	66.0(59.0–72.0)	<0.001^a
WBC ( × 10³/μL)	7.87 ± 3.14	8.05 ± 3.31	9.93 ± 5.0	10.04 ± 5.25	0.139
Creatinine(mg/dL)	0.78 ± 0.19	0.78 ± 0.16	0.80 ± 0.17	0.82 ± 0.19	0.813
ALT (U/L)	23.64 ± 10.25	24.64 ± 9.58	24.76 ± 10.24	30.64 ± 23.30	0.313
Albumin (g/dL)	3.88 ± 0.33	3.75 ± 0.25	3.74 ± 0.30	3.68 ± 0.22	0.070
INR	1.01 ± 0.08	1.01 ± 0.06	1.0 ± 0.06	1.04 ± 0.05	0.244
Cholesterol (mg/dL)	136.12 ± 17.48	139.72 ± 26.38	171.12 ± 29.35	192.28 ± 27.61	>0.001^a
LDL-C (mg/dL)	70.80 ± 8.01	73.92 ± 9.84	84.24 ± 10.66	95.0 ± 9.70	>0.001^a
HDL-C (mg/dL)	50.08 ± 5.60	48.0 ± 6.22	50.36 ± 6.20	49.36 ± 5.52	0.497
Triglycerides (mg/dL)	120.44 ± 10.88	122.08 ± 14.67	139.64 ± 20.38	161.68 ± 16.89	>0.001^a
hs-CRP (mg/dL)	9.98 ± 5.91	12.88 ± 8.90	51.44 ± 19.98	108.92 ± 40.90	<0.001^a
IL 6 (pg/mL)	2.79 ± 1.44	6.22 ± 1.75	16.07 ± 9.63	22.24 ± 9.17	>0.001^a
Serum ferritin (ng/mL)	90.0(58.0–150.0)	110.0(45.0–250.0)	199.0(110.0–369.0)	442.0(179.0–475.0)	<0.001^a
D-dimer (ng/mL) median (IQR)	270.0(225.0–512.0)	599.0(224.0–745.0)	775.0(692.0–884.0)	954.0(865.0–1134.0)	<0.001^a
HbA1c % (mmol/mol)	4.7 ± 0.4 (28)	4.8 ± 0.4 (29)	4.8 ± 0.4 (29)	5.0 ± 0.3 (31)	0.249
Admission FPG (mg/dL)	79.60 ± 7.21	90.08 ± 12.42	121.32 ± 18.91	134.28 ± 18.11	<0.001^a
Insulin level (mIU/mL)	4.09 ± 1.06	5.33 ± 1.52	22.51 ± 8.13	28.54 ± 8.03	<0.001^a
HOMA-IR	0.50(0.39 − 0.62)	0.65(0.53–0.88)	3.58(2.71–3.79)	5.15(3.71–5.38)	<0.001^a
QUICKI	0.39 ± 0.02	0.38 ± 0.02	0.30 ± 0.03	0.27 ± 0 .02	<0.001^a
FIRI	0.70(0.57–0.94)	0.98(0.84–1.36)	7.39(5.87–7.96)	9.5(7.96–9.55)	<0.001^a
TyG index	4.58 ± 0.06	4.65 ± 0.09	4.86 ± 0.13	4.95 ± 0.09	<0.001^a
FPG after 3 months (mg/dL)	81.28 ± 6.61	84.04 ± 7.79	112.33 ± 12.39	124.60 ± 5.27	<0.001^a
Cured No. (%)	25 (100)	25 (100)	12 (48)	5 (20%)	<0.01^a

Statistically significant at P ≤ 0.05.

FIRI, Fasting Insulin Resistance Index; FPG, fasting plasma glucose; HDL-C, high-density lipoprotein cholesterol; HOMA-IR, homeostatic model assessment-insulin resistance hs-CRP, high sensitivity C-reactive protein; IL-6, interleukin 6; LDL-C, low-density lipoprotein cholesterol; QUICKI, Quantitative Insulin Sensitivity Check Index; SD, standard deviation; TyG index, triglyceride-glucose index.

The total sample was reclassified according to outcome into two groups (cured or died). Comparison between both groups is shown in Table 2. Patients in the severe group (n = 25) were further subdivided into two subgroups [cured; 12 (48%) or died 13 (52%)]. Comparison between both subgroups revealed significantly higher admission FPG, fasting insulin, HOMA-IR, FIRI, and TyG index values in patients who died than those who cured (P = 0.004, 0.004, 0.001, 0.001, and 0.005, respectively). In contrast, QUICKI was significantly lower in died cases than in cured cases (P = 0.008). However, there was no significant difference between cured and died subgroups of severe group regarding demographic, inflammatory, hematological, or lipid parameters.

Table 2.

Comparison Between the Two Studied Groups According to Different Parameters in the Total Sample (n = 100)

Parameter mean ± SD	Cure (n = 67)	Died (n = 33)	P
Male No. (%)	35 (52.2)	17 (51.5)	0.964
Age (years)	47.18 ± 13.98	60.45 ± 12.64	0.001^a
WBC ( × 10³/μL)	8.33 ± 3.67	10.27 ± 5.29	0.035^a
Serum creatinine (mg/dL)	0.78 ± 0.16	0.82 ± 0.20	0.278
Serum albumin (g/dL)	3.80 ± 0.28	3.69 ± 0.29	0.090
INR	1.01 ± 0.07	1.02 ± 0.06	0.310
Serum total cholesterol (mg/dL)	148.18 ± 30.97	183.42 ± 28.46	>0.001^a
LDL-C (mg/dL)	76.13 ± 12.01	90.85 ± 10.53	>0.001^a
Triglycerides (mg/dL)	127.3 ± 20.49	153.7 ± 17.16	>0.001^a
hs-CRP (mg/dL)	24.41 ± 30.82	89.24 ± 41.98	>0.001^a
IL 6 (pg/mL)	7.11 ± 6.31	21.41 ± 10.01	>0.001^a
Serum ferritin (ng/mL)	130.0 (56.50–253.0)	220.0 (169.0–450.0)	>0.001^a
D-dimer (ng/mL)	569.0 (239.0–767.0)	897.0 (749.0–974.0)	>0.001^a
HbA1c % (mmol/mol)	4.9 ± 0.4 (30(	4.8 ± 0.3 (2.9)	0.560
Admission FPG (mg/dL)	91.22 ± 17.41	136.97 ± 12.16	>0.001^a
Serum insulin level (mIU/mL)	8.99 ± 9.12	27.56 ± 6.61	>0.001^a
HOMA-IR	0.66 (0.51–0.95)	3.94 (3.60–5.25)	>0.001^a
QUICKI	0.36 ± 0.04	0.28 ± 0.02	>0.001^a
FIRI	0.98 (0.71–1.60)	8.69 (7.86–9.50)	>0.001^a
TyG index	4.66 ± 0.13	4.96 ± 0.08	>0.001^a

Statistically significant at P ≤ 0.05.

Correlation between FPG, HOMA-IR, TyG index, and different parameters are shown in Table 3. Regression analysis was performed to identify factors affecting mortality in the total sample (Table 4). Univariate regression analysis revealed that admission FPG, HOMA-IR, FIRI, total cholesterol, LDL-C, triglycerides, D-dimer, IL6, hs-CRP, and ferritin were the independent risk factors predicting mortality, whereas admission FPG was the only independent risk factor predicting mortality in multivariate regression analysis in the total sample.

Table 3.

Correlation Between Admission Fasting Plasma Glucose, Homeostatic Model Assessment-Insulin Resistance, Triglyceride-Glucose Index, and Different Parameters for Total Sample (n = 100)

	Admission fasting plasma glucose (mg/dL)		HOMA-IR		TyG index
	r_s	P	r_s	P	r_s	P
Age (years)	0.444^a	<0.001^a	0.627^a	<0.001^a	0.552^a	<0.001^a
Hemoglobin (g/dL)	0.115	0.254	−0.108	0.284	−0.124	0.218
WBCs ( × 10³/μL)	0.275^a	0.006^a	0.280^a	0.005^a	0.311^a	0.002^a
Platelets ( × 10³/μL)	0.007	0.942	−0.069	0.496	−0.018	0.857
Blood urea (mg/dL)	0.175	0.081	0.110	0.274	0.196	0.051
Serum creatinine (mg/dL)	0.096	0.341	0.143	0.155	0.117	0.246
ALT (U/L)	0.159	0.114	0.099	0.326	0.201^a	0.045^a
AST (U/L)	0.168	0.095	0.079	0.435	0.189	0.060
Serum albumin (g/dL)	0.178	0.077	−0.246^a	0.014^a	−0.249^a	0.012^a
Prothrombin activity	0.163	0.105	0.067	0.508	0.136	0.178
INR	0.095	0.345	0.220	0.028	0.161	0.110
Serum total cholesterol (mg/dL)	0.538^a	<0.001^a	0.679^a	<0.001^a	0.727^a	<0.001^a
LDL-C (mg/dL)	0.563^a	<0.001^a	0.672^a	<0.001^a	0.723^a	<0.001^a
HDL-C(mg/dL)	0.032	0.754	−0.019	0.853	0.019	0.849
Triglycerides (mg/dL)	0.636^a	<0.001^a	0.742^a	<0.001^a	0.843^a	<0.001^a
Hs-CRP (mg/dL)	0.685^a	<0.001^a	0.825^a	<0.001^a	0.671^a	<0.001^a
IL 6 (pg/mL)	0.743^a	<0.001^a	0.756^a	<0.001^a	0.791^a	<0.001^a
Serum ferritin (ng/mL)	0.434^a	<0.001^a	0.522^a	<0.001^a	0.499^a	<0.001^a
D-dimer (ng/mL)	0.592^a	<0.001^a	0.585^a	<0.001^a	0.588^a	<0.001^a
HbA1c % (mmol/mol)	0.038	0.709	0.110	0.274	0.090	0.371
HOMA-IR	0.857^a	<0.001^a	—	—	0.832^a	<0.001^a
Admission FPG (mg/dL)	—	—	0.799^a	<0.001^a	0.936^a	<0.001^a
TyG index	0.936^a	<0.001^a	0.832^a	<0.001^a	—	—
QUICKI	0.886^a	<0.001^a	−0.989^a	<0.001^a	−0.899^a	<0.001^a
FIRI	0.908^a	<0.001^a	0.976^a	<0.001^a	0.916^a	<0.001^a
FPG after 3 months (mg/dL)	0.611^a	<0.001^a	0.490^a	<0.001^a	0.619^a	<0.001^a

Statistically significant at P ≤ 0.05.

r, Pearson coefficient.

Table 4.

Univariate and Multivariate Analysis for the Parameters Affecting Mortality Cases for Total Sample (n = 25)

Total sample	Univariate		Multivariate
Total sample	P	OR (95% CI)	P	OR (95% CI)
Admission FPG (10 mg/dL)	<0.001^a	1.286^a (1.153–1.433)	0.027^a	1.708^a (1.063–2.745)
HOMA-IR	<0.001^a	3.497^a (2.229–5.485)	0.417	0.282 (0.013–6.004)
Serum total cholesterol (mg/dL)	<0.001^a	1.035^a (1.019–1.052)	0.935	1.007 (0.842–1.205)
LDL-C (mg/dL)	<0.001^a	1.107^a (1.060–1.156)	0.422	1.223 (0.749–1.996)
Triglycerides (mg/dL)	<0.001^a	1.064^a (1.037–1.093)	0.513	0.970 (0.886–1.062)
FIRI	<0.001^a	2.672^a (1.629–4.383)	0.339	2.806 (0.338–23.305)
D-dimer (ng/mL)	<0.001^a	1.004^a (1.002–1.005)	0.806	0.999 (0.995–1.004)
Hs-CRP	<0.001^a	1.048^a (1.029–1.067)	0.051	1.086 (0.999–1.181)
IL 6 (pg/mL)	<0.001^a	1.196 (1.119–1.278)	0.511	0.943 (0.791–1.124)
Serum ferritin (ng/mL)	0.001^a	1.005^a (1.002–1.008)	0.213	0.995 (0.986–1.003)

Statistically significant at P ≤ 0.05.

CI, confidence interval; OR, odds ratio.

The ROC curve was used to identify cutoff values of admission FPG, FIRI, and HOMA-IR that predict mortality in the total sample (Fig. 1). Regarding admission FPG, a value >117 mg/dL was able to predict mortality (P < 0.001) with an excellent sensitivity, specificity, and AUC (96.97, 86.57, and 0.968, respectively). Moreover, HOMA-IR value >2.2 was able to predict mortality in the total sample (P < 0.001) with very high sensitivity, specificity, and AUC (96.97, 82.09, and 0.930, respectively), whereas FIRI value >6.33 was able to predict mortality (P < 0.001) with very high sensitivity, specificity, and AUC (97, 85.07, and 0.959*, respectively).

FIG. 1.

ROC curve for different parameters to predict mortality for the total sample (n = 100). ROC, receiver operating characteristic.

Discussion

In the era of COVID-19 pandemic, a bidirectional relationship exists between DM and COVID-19 infection. This depends on various factors. However, hyperglycemia per se without pre-existing DM is accompanied by worse COVID-19 infection outcome.

In this study, the mean FPG at admission was significantly higher in severe and critical cases than mild and moderate cases with increased tendency of admission FPG with increased severity of infection. It was also significantly higher in patients who died than cured cases.

This study also showed a significantly higher FPG after 3 months in cured severe and critical cases than in mild and moderate cases where FPG level increased with increasing severity of infection. In addition, there was a highly significant positive correlation between FPG at admission and in follow-up after 3 months. Admission FPG also showed a significant positive correlation with various inflammatory, IR, and lipid parameters. This highlights the close association between admission FPG and outcome of COVID-19 infection. This may help in predicting the high-risk cases of severe disease and death at admission.

In agreement with the results of this study, Bode et al.¹ reported that hyperglycemia in COVID-19 infection could increase mortality risk. Yang et al.¹⁹ also reported a relation of COVID-19-associated morbidity and mortality with high FPG or acute uncontrolled hyperglycemia. Moreover, Reiterer et al.²⁰ found that patients with hyperglycemia have worse outcome with higher prevalence of hyperglycemia in ARDS patients. Furthermore, Yang et al.,²¹ Lazarus et al.,²² and Sachdeva et al.²³ found that higher admission blood glucose in patients with COVID-19 infection was associated with critical and fatal disease.

Mamtani et al.² reported that patients without diabetes who had hyperglycemia at admission by COVID-19 infection had poor clinical outcome and were at higher risk of mortality. They concluded that hyperglycemia within the first 48 hr of admission was an independent prognostic factor for COVID-19-infected patients. Moreover, in this study, admission FPG was the only independent predictor of mortality in multivariate regression analysis although many other factors independently predicted mortality in univariate analysis. With every 10 mg/dL increment in FPG there is 1.7-fold increase risk of mortality. Furthermore, a cutoff value of admission FPG >117 mg/dL was a predictor of mortality with excellent sensitivity, specificity, and AUC.

Coppelli et al.,⁴ in agreement with the results of this study, revealed that hyperglycemia accompanying admission in patients with COVID-19 infection was the only independent factor predicting disease course with 30% increased mortality risk even when compared with patients with DM. Alshukry et al.²⁴ showed similar results with 1.52 increased risk of mortality with each 18 mg/dL increase in FPG level. Lazarus et al.²² also reported that each 18 mg/dL rise in FPG was associated with 33% risk of severe disease. Wu et al.²⁵ observed that admission blood glucose level was an independent predictor of mortality and correlated with the severity of infection in critical cases but not in noncritical cases.

However, in contrast, few studies^26
–28 reported no relation between admission blood glucose and disease prognosis. This discordance between their results and the results of this study arise from the difference in patients' characteristics and study design.

This study found that inflammatory markers (hs-CRP, ferritin, IL-6, and D-dimer) were significantly higher in severe and critical cases than in mild and moderate cases. They also showed significantly higher values in died than cured cases. Moreover, these inflammatory markers showed significant positive correlation with admission FPG level, TyG index, and HOMA-IR.

In consistence with the results of this study, Faraj and Jalal²⁹ found a significantly higher level of inflammatory markers (CRP, ferritin, and IL6) in severe cases than in mild cases infected with COVID-19. They also reported a relation between severity of disease progression and the serum level of these inflammatory markers. Ren et al.³⁰ found significantly higher FPG, CRP, ferritin, TNF-α, and WBCs in severe cases compared with mild cases although there was no significant difference between the two groups regarding HbA1c.

This study was the only one to discuss the relation between four IR markers and COVID-19 infection severity and outcome and identified the cutoff values of IR parameters that predict mortality. The results showed significantly higher HOMA-IR, TyG index, and FIRI and a significantly lower QUICKI in patients with severe and critical cases compared with mild and moderate cases. In addition, the same results were shown between died and cured cases in the total sample (n = 100) and in severe group (n = 25). HOMA-IR, TyG index, and admission FPG showed significant positive correlations with follow-up FPG, lipid, inflammatory, and other IR parameters. Furthermore, HOMA-IR and FIRI were identified as independent predictors of mortality in univariate but not multivariate.

In this study, HOMA-IR and FIRI values exceeding 2.2 and 6.33, respectively, were able to predict mortality with excellent sensitivity, specificity, and AUC. The adipose tissue dysfunction is the base of IR, which in turn causes hyperglycemia that is considered the main predictor of patients' outcome.²⁰

Gojda et al.³¹ similarly found a higher IR in the acute phase of COVID-19 infection, but it did not persist in the follow-up period. Chang et al.³² studied the relation between TyG index and severity of COVID-19 infection. They found higher mortality and increased infection severity with higher TyG index value and they concluded that TyG index could be used as a severity predictor in patients with COVID-19 infection. Liontos et al.³³ reported that IR indices, inflammatory markers (IL6 and CRP), and hyperglycemia were associated with prolonged hospital stay, intubation, and death. Reiterer et al.²⁰ suggested that hyperglycemia present in severe cases is caused mainly by IR. They also related IR in patients with COVID-19 infection to the direct effect of the viral infection to adipose tissue.

Conclusions

Admission FPG is a good predictor of disease severity, patient outcome, and risk of mortality in patients admitted with COVID-19 infection. In patients without pre-existing DM, acute hyperglycemia is a better predictor of poor outcome than HbA1c. This study highlighted the association of hyperglycemia and its related IR in predicting the severity of COVID-19 infection and outcome.

The cutoff values of admission FPG, HOMA-IR, and FIRI predicted mortality in the studied cohort with excellent sensitivity, specificity, and AUC. These results will help in identifying high-risk patients at admission and will allow proper management as early as possible to decrease worse outcome.

However, the study has some limitations. First the follow-up for cured patients is required to be extended to evaluate patients who will develop DM in the future. Second, the role of drugs that improve IR on patients' outcome needs further studies.

Footnotes

Acknowledgments

The authors dedicate this study to the soul of their dear professor, Dr. Mohammed Ahmed Badr, professor of diabetes and metabolism, Internal Medicine Department, Faculty of Medicine, Alexandria University (May his soul rest in peace) for his help and effort at the beginning of this study. They would also like to thank Diabetes and Metabolism Unit, Internal Medicine Department, Faculty of Medicine, Alexandria University, and Alexandria Fever Hospital.

Authors' Contributions

Conceptualization, methodology, supervision, and writing—review and editing by F.Z.A., M.H.M., and M.H.B. Investigation, methodology, supervision, and writing—review and editing by S.A.I. Data curation, formal analysis, investigation, resources, and writing—review and editing by H.M.Z. Methodology, resources, supervision, and writing—original draft by H.S.K.

Data Availability

The data sets of this study are available on reasonable request from the corresponding author.

AI Statement

AI has not been used in the writing process.

Author Disclosure Statement

No conflicting financial interests exist.

Funding Information

No funding was received for this article.

References

Bode

, Garrett

, Messler

, et al. Glycemic characteristics and clinical outcomes of COVID-19 patients hospitalized in the United States. J Diabetes Sci Technol, 2020; 14(4):813–821; doi: 10.1177/1932296820924469

Mamtani

, Athavale

, Abraham

, et al. Association of hyperglycaemia with hospital mortality in nondiabetic COVID-19 patients: A cohort study. Diabetes Metab, 2021; 47(3):101254; doi: 10.1016/j.diabet.2021.101254

Singh

, Singh

. Hyperglycemia without diabetes and new-onset diabetes are both associated with poorer outcomes in COVID-19. Diabetes Res Clin Pract, 2020; 167:108382; doi: 10.1016/j.diabres.2020.108382

Coppelli

, Giannarelli

, Aragona

, et al. Hyperglycemia at hospital admission is associated with severity of the prognosis in patients hospitalized for COVID-19: The Pisa COVID-19 Study. Diabetes Care, 2020; 43(10):2345–2348; doi: 10.2337/dc20-1380

Hussain

, Bhowmik

, do Vale Moreira

. COVID-19 and diabetes: Knowledge in progress. Diabetes Res Clin Pract, 2020; 162:108142; doi: 10.1016/j.diabres.2020.108142

Donath

, Dinarello

, Mandrup-Poulsen

. Targeting innate immune mediators in type 1 and type 2 diabetes. Nat Rev Immunol, 2019; 19(12):734–746; doi: 10.1038/s41577-019-0213-9

Pucino

, Bombardieri

, Pitzalis

, et al. Lactate at the crossroads of metabolism, inflammation, and autoimmunity. Eur J Immunol, 2017; 47(1):14–21; doi: 10.1002/eji.201646477

Eketunde

, Mellacheruvu

, Oreoluwa

. A review of postmortem findings in patients with COVID-19. Cureus, 2020; 12(7):e9438; doi: 10.7759/cureus.9438

Alexandraki

, Piperi

, Ziakas

, et al. Cytokine secretion in long-standing diabetes mellitus type 1 and 2: Associations with low-grade systemic inflammation. J Clin Immunol, 2008; 28(4):314–321; doi: 10.1007/s10875-007-9164-1

10.

Finucane

, Davenport

. 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

11.

Santos

, Magro

, Evangelista-Poderoso

, et al. Diabetes, obesity, and insulin resistance in COVID-19: Molecular interrelationship and therapeutic implications. Diabetol Metab Syndr, 2021; 13(1):23; doi: 10.1186/s13098-021-00639-2

12.

World Health Organization. Director-General's remarks at the media briefing on 2019-nCoV on 11 February 2020. World Health Organization; 2020.

13.

Bergenstal

, Beck

, Close

, et al. Glucose management indicator (GMI): A new term for estimating A1C from continuous glucose monitoring. Diabetes Care, 2018; 41(11):2275–2280; doi: 10.2337/dc18-1581

14.

Salgado

, Carvalho

, Oliveira

, et al. Insulin resistance index (HOMA-IR) in the differentiation of patients with non-alcoholic fatty liver disease and healthy individuals. Arq Gastroenterol, 2010; 47(2):165–169; doi: 10.1590/s0004-28032010000200009

15.

Reaven

, Abbasi

, McLaughlin

. Obesity, insulin resistance, and cardiovascular disease. Recent Prog Horm Res, 2004; 59:207–223; doi: 10.1210/rp.59.1.207

16.

Petersen

, Shulman

. Mechanisms of insulin action and insulin resistance. Physiol Rev, 2018; 98(4):2133–2223; doi: 10.1152/physrev.00063.2017

17.

Fan

, Miller

, Lee

, et al. Diagnosing sepsis—The role of laboratory medicine. Clin Chim Acta, 2016; 460:203–210; doi: 10.1016/j.cca.2016.07.002

18.

Spring

, Winkler

, Levy

. The influence of various patient characteristics on D-dimer concentration in critically ill patients and its role as a prognostic indicator in the intensive care unit setting. Clin Lab Med, 2014; 34(3):675–686; doi: 10.1016/j.cll.2014.06.015

19.

Yang

, Yu

, Xu

, et al. Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: A single-centered, retrospective, observational study. Lancet Respir Med, 2020; 8(5):475–481; doi: 10.1016/S2213-2600(20)30079-5

20.

Reiterer

, Rajan

, Gomez-Banoy

, et al. Hyperglycemia in acute COVID-19 is characterized by insulin resistance and adipose tissue infectivity by SARS-CoV-2. Cell Metabol, 2021; 33:2174–2188; doi: 10.1016/j.cmet.2021.09.009.e5

21.

Yang

, Cai

, Zhang

. Hyperglycemia at admission is a strong predictor of mortality and severe/critical complications in COVID-19 patients: A meta- analysis. Biosci Rep, 2021; 41(2):BSR20203584; doi: 10.1042/BSR20203584

22.

Lazarus

, Audrey

, Wangsaputra

, et al. High admission blood glucose independently predicts poor prognosis in COVID-19 patients: A systematic review and dose-response meta-analysis. Diabetes Res Clin Pract, 2021; 171:108561; doi: 10.1016/j.diabres.2020.108561

23.

Sachdeva

, Desai

, Gupta

, et al. Admission hyperglycemia in non-diabetics predicts mortality and disease severity in COVID-19: A pooled analysis and meta-summary of literature. SN Compr Clin Med, 2020; 2(11):2161–2166; doi: 10.1007/s42399-020-00575-8

24.

Alshukry

, Bu Abbas

, Ali

, et al. Clinical characteristics and outcomes of COVID-19 patients with diabetes mellitus in Kuwait. Heliyon, 2021; 7(4):e06706; doi: 10.1016/j.heliyon.2021.e06706

25.

, Tang

, Cheng

. Diabetes increases the mortality of patients with COVID-19: A meta-analysis. Acta Diabetol, 2021; 58(2):139–144; doi: 10.1007/s00592-020-01546-0

26.

Kim

, Jeon

, Kim

, et al. The clinical characteristics and outcomes of patients with moderate-to-severe coronavirus disease 2019 infection and diabetes in Daegu, South Korea. Diabetes Metab J, 2020; 44(4):602–613; doi: 10.4093/dmj.2020.0146

27.

Mehta

, Kohn

, Koliwad

, et al. Lack of association between either outpatient or inpatient glycemic control and COVID-19 illness severity or mortality in patients with diabetes. BMJ Open Diabetes Res Care, 2021; 9(1):e002203; doi: 10.1136/bmjdrc-2021-002203

28.

Koh

, Moh

AMC

, Yeoh

, et al. Diabetes predicts severity of COVID-19 infection in a retrospective cohort: A mediatory role of the inflammatory biomarker C-reactive protein. J Med Virol, 2021; 93(5):3023–3032; doi: 10.1002/jmv.26837

29.

Faraj

, Jalal

. IL1β, IL-6, and TNF-α cytokines cooperate to modulate a complicated medical condition among COVID-19 patients: Case-control study. Ann Med Surg (Lond), 2023; 85(6):2291–2297; doi: 10.1097/MS9.0000000000000679

30.

Ren

, Yang

, Wang

, 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

31.

Gojda

, Koudelková

, Ouřadová

, et al. Severe COVID-19 associated hyperglycemia is caused by beta cell dysfunction: A prospective cohort study. Nutr Diabetes, 2023; 13(1):11; doi: 10.1038/s41387-023-00241-7

32.

Chang

, Jeon

, Song

, et al. Association of triglyceride-glucose index with prognosis of COVID-19: A population-based study. J Infect Public Health, 2022; 15(8):837–844; doi: 10.1016/j.jiph.2022.06.014

33.

Liontos

, Biros

, Kavakli

, et al. Glycemic dysregulation, inflammation and disease outcomes in patients hospitalized with COVID-19: Beyond diabetes and obesity. Viruses, 2023; 15(7):1468; doi: 10.3390/v15071468