The Chinese Center for Disease Control and Prevention had reported that since the pandemic control measures of coronavirus disease 2019 (COVID-19) were fully lifted in mainland China in 12/2022, the peak number of COVID-19 nucleic acid-positive cases had reached 6.94 million, admissions to hospitals had reached a peak of 1.625 million, of which the highest number of severe cases had reached 128 thousand, and the cumulative number of deaths had reached 4273 by January 2023.

Diabetes has already become the second most common comorbidity of COVID-19 due to the coinciding of two global pandemics (1, 2). A meta-analysis including 7 studies with 1,576 patients showed the prevalence of diabetes of approximately 9.7% (95% CI: 7.2–12.2%) (3). Another meta-analysis was a comprehensive systematic search including data from 76,993 patients (4). According to this study, the prevalence of diabetes was estimated to be 7.87% (95% CI: 6.57–9.28%). Poor glycemic control increased the risk of mortality, morbidity, and secondary infections (5, 6).

These associations between diabetes and worse outcomes in COVID-19 patients were incontrovertible, as blood glucose fluctuation was not conducive to the improvement of disease, and inflammation caused by hyperglycemia led to increased mortality (7, 8). However, excessively tight glycemic control may increase the risk of hypoglycemia, which also increased mortality (9).The impact of COVID-19 on the patients and the use of glucocorticoids and nutritional support during the treatment increased blood glucose fluctuations, which had adverse effects on the prognosis (10). The UK Diabetes guidelines recommend a blood glucose target of 110 to 180mg/dL for diabetes patients with COVID-19, and a blood glucose level of less than 220mg/dL for patients with hypoglycemia and high risk factors (including the elderly, patients with low body weight, patients with severe COVID-19 and/or renal impairment) (11). American Diabetes Association guidelines recommend targeting blood glucose < 180 mg/dL in critically ill patients (12). Clinicians face a significant challenge in improving outcomes for individuals with COVID-19 and type 2 diabetes(T2D) due to uncertainty surrounding the optimal degree of glycemic management and its potential impact on treatment benefits and risks. The definition of optimal blood glucose control remains controversial (13). The wide application of hormonal and nutritional support treatment has led to significant fluctuations in blood glucose levels in clinical practice, making it challenging to maintain the general range. Consequently, our study aimed to analyze glycemic profiles using intermittently scanned continuous glucose monitoring (isCGM) to determine a more clinically practical threshold for TIR and investigate its correlation with prognosis.

In our observational study, data of patients admitted to Qilu Hospital of Shandong University (public tertiary care) from Dec 2022 to Apr 2023 were analyzed. The patients all had moderate or severe cases and were diagnosed according to the guidelines issued by the World Health Organization (WHO) (14), meeting at least the following criteria: positive COVID-19 RNA PCR and characteristic imaging manifestations of novel coronavirus pneumonia. Additionally, eligible patients were those with T2D that had been diagnosed before admission or with newly diagnosed T2D after admission. All patients met the diagnostic criteria of T2D: typical diabetes symptoms plus random blood glucose ≥11.1 mmol/L or plus fasting plasma glucose(FPG)≥7.0 mmol/L or OGTT 2h blood glucose≥11.1 mmol/L. Exclusion criteria included patients who were intubated on admission and those younger than 18 years of age. The study was approved by the Ethics Committees of Qilu Hospital of Shandong University (KYLL-202307–047). Trial Registration: clinicaltrials.gov Identifier: NCT06156137 (Registered November 24, 2023).

Patient information that we collected through electronic medical records include gender, age, vital signs, symptoms on admission, duration of diabetes, comorbidities, FPG, hemoglobin A_1c (HbA_1c), alanine aminotransferase (ALT), aspartate aminotransferase (AST), total protein (TP), total cholesterol (TC), triglycerides (TG), serum creatinine, uric acid, estimated glomerular filtration rate (eGFR), inflammatory biomarkers, brain natriuretic peptide (BNP), CK-MB and medication, including oral hypoglycemic agent (OHA), insulin, anticoagulant drugs and glucocorticoids. CGM was initiated on admission. Diabetic meals were ordered for all patients during hospitalization.

All patients were equipped with isCGM sensors (FreeStyle Libre Flash glucose monitoring system; Abbott Diabetes Care Ltd, UK) on admission, and the nurse retrieved the probe when the patient was discharged or when the composite endpoint was reached. The routine protocol for glucose monitoring during hospitalization was fixed at four swipes daily (fasting, premeal and bedtime). In addition, scans can be performed when the patient encounters symptoms of hypoglycemia or any other discomfort. Measures of glycemic variability, such as time-in-range (TIR), time-below-range (TBR) and time-above-range (TAR), mean sensor glucose level and coefficient of variation (CV) of glucose levels, were calculated from isCGM records. TIR was defined as the percentage of time within the following ranges: 70–180 mg/dL, 80–190 mg/dL, 90–200 mg/dL, 100–210 mg/dL, 110–220 mg/dL, and 120–230 mg/dL.

A composite adverse outcome included a hospital stay of more than 20 days, admission to the intensive care unit, the need for mechanical ventilation, and death.

Data from this cross-sectional study showed that optimal glycemic control during hospitalization was associated with a lower risk of severe illness and death in patients with COVID-19. After adjusting for covariates, maintaining TIR within the thresholds of 80 to 190 mg/dL significantly relates to favorable outcomes.

In our study, the patient population was divided into two cohorts based on the occurrence of composite adverse events. The proportion of severe COVID-19 cases at admission was higher in the population with composite adverse events than in the second cohort (63% vs. 33.3%, P = 0.002). Although patients with composite adverse outcomes were more likely to be male and older than 65 years with comorbidities and higher levels of inflammatory, endothelial, and coagulopathy markers on admission, there was no significant difference between the two groups. Patients achieving composite adverse outcomes had significantly higher CK-MB and LDH levels on admission. When analyzing TIR as a factor influencing outcome, all of the above confounding variables were adjusted for to reach the following conclusion: TIR values with thresholds of 80 to 190 mg/dL were significantly associated with a lower risk of the composite adverse outcomes.

Previous studies have shown that variability is a potential risk predictor of death and other complications (4, 15). The presence of COVID-19 has been shown to play a significant role in impairing blood glucose control within the range of 70–150 mg/dL (13). A study of 548 patients with COVID-19 and T2D has confirmed that the parameters such as mean glucose, peak glucose, and the magnitude of glycemic fluctuations in the early stage of hospitalization are significantly correlated with adverse outcomes, and are closely related to increased hospitalization expenses, prolonged hospitalization time, and increased risk of all-cause death (16). A small-sample study (17) suggested that maintaining TIR (70–160 mg/dL) >70% could improve outcomes. In clinical practice, we found that only 15.87% of patients achieved that target, and the average TIR in our study was 39.36% during the pandemic. Inpatient medication (corticosteroids) and enteral and parenteral nutrition contribute to hyperglycemia (18). The widespread use of glucocorticoids caused patients to experience wide fluctuations in blood glucose levels, which may have more adverse effects than sustained hyperglycemia. In our study, more than 70% of the patients were received glucocorticoids therapy, and 75% were treated with enteral or parenteral nutrition, which resulted in a high mean sensor glucose level [203.57 mg/dL (162.7–235.88)] and a wide CV of glucose values [33.29% (27.88 to 37.62)]. This also explained why the TIR threshold of COVID-19 patients with T2D was higher.

Moreover, the elevation of cortisol levels resulting from COVID-19 infection, stress, and similar factors can contribute to excessive hepatic gluconeogenesis, impaired glucose utilization, and insulin deficiency (19–21). There is a suggested direct impact of SARS-CoV-2 on pancreatic β-cell function and survival, exacerbating rapid and severe metabolic deterioration in individuals with preexisting diabetes (22, 23). Angiotensin-converting enzyme 2 (ACE 2) potentially serves as a crucial molecular link between COVID-19 severity and insulin resistance (23–25). Our findings supported this hypothesis, as the patients who achieved the composite adverse outcomes had a significantly lower TIR (80–190 mg/dL) and a higher TAR >190 mg/dL. Furthermore, they used a higher maximum insulin dose during hospitalization [34(18–47) vs. 19(0–40), P = 0.046]. In this study, we found that poor glycemic control was associated with a worse outcome that included a higher need for medical intervention, hospitalization, and mortality. The insights gained here provide direct suggestions for the clinical management of T2D during the COVID-19 pandemic.

Excessive glycemic control leading to severe hypoglycemia has been associated with increased mortality rates (9). The international consensus on TIR (26) indicated that although evidence regarding TIR for older or high-risk individuals is limited, several studies have demonstrated an elevated risk for hypoglycemia. Therefore, they reduced the TIR target from 70% to 50%. In our study, the age of enrolled patients was relatively high, the mean age was 71.59 ± 12.24 years old, and the TIR (80 to 190 mg/dL) corresponded to 54.73% and had a maximum Youden index. This cutoff value had good clinical significance.

The major advantage of our study lies in the utilization of the isCGM system for T2D patients complicated with COVID-19, enabling comprehensive assessment of hyperglycemia, hypoglycemia, and glycemic variability. Our study has presented the appropriate threshold and cutoff point for TIR in patients with COVID-19 and T2D, which is more relevant to clinical practice. However, several limitations need to be acknowledged. Firstly, it was a retrospective study, which may introduce patient selection bias. Secondly, the sample size was relatively modest and might not fully capture the complexity of the general population. Therefore, large-scale prospective cohort studies involving ethnically diverse cohorts from different geographical regions are warranted to gain a better understanding of the association between glycemic control and COVID-19 progression. Finally, it should be noted that our analysis excluded individuals with type 1 diabetes, but glycemic control could also influence their outcomes.

In conclusion, maintaining a TIR (80–190 mg/dL) above 54.73% independently correlates with a significant reduction in composite adverse outcomes associated with COVID-19 infection among patients with T2D. These findings provide valuable insights into the clinical characteristics of glycemic variability in individuals affected by both COVID-19 and T2D while offering potential avenues for improving disease outcomes.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by the Ethics Committees of Qilu Hospital of Shandong University. The studies were conducted in accordance with the local legislation and institutional requirements. The medical records or biological specimens used in this study are obtained from previous clinical treatment, and the risk to the subjects is not greater than the minimum risk, and the exemption from informed consent will not adversely affect the rights and health of the subjects. Subjects’ privacy and personally identifiable information will be protected by the researchers. The project leader solemnly promises not to use the medical records and specimens that patients have explicitly refused to use before. This research project does not involve personal privacy and commercial interests, and the samples and related information are only used for this research project.

KT: Data curation, Formal analysis, Methodology, Supervision, Validation, Visualization, Writing – review & editing. RC: Methodology, Formal analysis, Writing – original draft. JBZ: Methodology, Formal analysis, Supervision, Writing – review & editing. LX: Formal analysis, Writing – original draft. YZ: Data curation, Writing – review & editing. HW: Data curation, Writing – original draft. JW: Data curation, Writing – original draft. WW: Data curation, Writing – original draft. YD: Data curation, Writing – original draft. HL: Data curation, Writing – original draft. JZ: Data curation, Writing – original draft.