Abstract
Aim
Poor outcomes of coronavirus disease 2019 (COVID‐19) have been linked to diabetes, but its relation to pre‐infection glycaemic control is still unclear.
Materials and Methods
To address this question, we report here the association between pre‐infection Haemoglobin A1c (HbA1c) levels and COVID‐19 severity as assessed by need for hospitalization in a cohort of 2068 patients with diabetes tested for COVID‐19 in Leumit Health Services (LHSs), Israel, between 1 February and 30 April 2020. Using the LHS‐integrated electronic medical records system, we were able to collect a large amount of clinical information including age, sex, socio‐economic status, weight, height, body mass index, HbA1c, prior diagnosis of ischaemic heart disease, depression/anxiety, schizophrenia, dementia, hypertension, cerebrovascular accident, congestive heart failure, smoking, and chronic lung disease.
Results
Of the patients included in the cohort, 183 (8.85%) were diagnosed with COVID‐19 and 46 were admitted to hospital. More hospitalized patients were female, came from higher socio‐economic background and had a higher baseline HbA1c. A prior diagnosis of cerebrovascular accident and chronic lung disease conferred an increased risk of hospitalization but not obesity or smoking status. In a multivariate analysis, controlling for multiple prior clinical conditions, the only parameter associated with a significantly increased risk for hospitalization was HbA1c ≥ 9%.
Conclusion
Using pre‐infection glycaemic control data, we identify HbA1c as a clear predictor of COVID‐19 severity. Pre‐infection risk stratification is crucial to successfully manage this disease, efficiently allocate resources, and minimize the economic and social burden associated with an undiscriminating approach.
Keywords: coronavirus infectious disease 2019 (COVID‐19), pre‐infection glycaemic control
1. INTRODUCTION
Poor outcomes of coronavirus disease 2019 (COVID‐19) have been linked to diabetes, 1 , 2 , 3 , 4 but its relation to pre‐infection glycaemic control is still unclear. Glycaemic levels during hospitalization have been linked to poor outcomes of COVID‐19. In a study of 28 patients with COVID‐19 and diabetes in Wuhan, China, the Haemoglobin A1c (HbA1c) was similar between 14 patients who were admitted to the intensive care unit (ICU) and those who were not. 5 In another study of 269 patients with severe disease upon hospitalization, hyperglycaemia was not identified as associated with severity of disease upon admission but was associated with death. 6 In a retrospective study of 201 patients with COVID‐19, pneumonia‐increased glucose levels were identified as a risk factor for the development of acute respiratory distress syndrome (ARDS) but not for death. 7 A similar observation was noted in a study of 85 patients hospitalized due to COVID‐19—admission glucose level was the strongest predictor of radiographic findings of ARDS. 8 In a large multi‐centred retrospective study including 7337 cases of COVID‐19 of which 952 patients had type 2 diabetes mellitus (T2DM), mean glycaemic levels between 70 and 180 mg/dl (3.9–10 mmol/L) during hospitalization were found to predict lower mortality. 9
Despite this emerging evidence, the relation between pre‐infection glycaemic control and COVID‐19 severity is still unknown. To address this question, we report here the association between pre‐infection HbA1c levels and COVID‐19 severity as assessed by need for hospitalization in a cohort of 2068 patients with diabetes tested for COVID‐19 in Leumit Health Services (LHSs), Israel, between 1 February and 30 April 2020.
2. METHODS
2.1. Population and clinical information
LHS is a large health maintenance organization (HMO) in Israel, which provides services to around 7,30,000 members nationwide. The comprehensive LHS electronic medical records (LHS‐EMR) system integrates medical, pharmaceutical, and laboratory information and is continuously updated. Using the LHS‐EMR, we identified patients with a prior diagnosis of diabetes (based on the Israeli diabetes registry 10 ) who underwent at least one test for COVID‐19 between 1 February and 30 April 2020. COVID‐19 testing was done only by physician referral (based on clinical criteria of exposure to confirmed COVID‐19 patients and/or symptoms suggesting COVID‐19) using the AllplexTM 2019‐nCoV Assay (Seegene Inc) until 10 March 2020, and since then—the COBAS SARS‐Cov‐2 6800/8800 (Roche Pharmaceuticals). Only patients with a positive COVID‐19 test were considered as having COVID‐19. Other information collected from LHS‐EMR included age, sex, socio‐economic status (SES—defined according to the patients home address, using the Israeli Central Bureau of Statistics classification 11 ), weight, height, body mass index (BMI), HbA1c, prior diagnosis of ischaemic heart disease (IHD), depression/anxiety, schizophrenia, dementia, hypertension, cerebrovascular accident (CVA), congestive heart failure (CHF), smoking, and chronic lung disease. The validity of the diagnoses in the registry was previously examined and confirmed as high. 12 Among the selected population, we identified patients who contracted COVID‐19 and those who were hospitalized due to COVID‐19. In Israel, most COVID‐19‐positive patients were treated as outpatients in what is referred to as home hospitalization. This was done either in the patients' homes or converted hotels. LHS home‐hospitalized patients received a pulse oxymeter and were followed twice daily by nurses via telemedicine or phone. Nurses were instructed to immediately refer a patient to a physician if his fever was ≥38°C, blood oxygen saturation was ≤95%, or shortness of breath developed. Hospitalization was at the discretion of the treating physician. The study protocol was approved by the Shamir Medical Center Review Board and the Research Committee of LHS.
2.2. Statistical analysis
Statistical analysis was conducted using a STATA 12 software (StataCorp LP). Assumptions were two sided with an α of 0.05. Initial analysis compared demographic characteristics between the patients with positive or negative Covid‐19 tests, using Student's t‐test and Fischer's exact χ 2 test for continuous and categorical variables, respectively, based on normal distribution and variable characteristics. Categorical data are shown in counts and percentages. Data on continuous variables with normal distribution are presented as means with 95% confidence intervals (CI). Preliminary evaluation of risk estimates was conducted by stratified analyses. HbA1c was stratified as: <7% (53 mmol/mol); 7%–7.9% (53–62.8 mmol/mol); 8%–8.9% (63.9–73.8 mmol/mol); or ≥9% (74.9 mmol/mol), and obesity was defined as a BMI ≥ 30 kg/cm2. Only patients who had at least one HbA1c test in the year prior to the COVID‐19 test were included in the analysis. In patients who had more than one HbA1c measurement during the year prior to COVID‐19 testing, the most recent measurement was used. Subsequently, multivariate logistic regression was used to estimate the odds ratios (OR) and 95% CI for the independent association between the following clinical characteristics—age, HbA1c ≥ 9%, sex, SES, smoking, IHD, depression/anxiety, schizophrenia, dementia, hypertension, CVA, CHF, smoking, chronic lung disease, obesity, and COVID‐19 disease severity has assessed by the need for hospitalization.
3. RESULTS
3.1. Clinical characteristics of patients with diabetes and COVID‐19
We identified 2068 patients with diabetes who were tested for COVID‐19 (ages 14–103 years) of which 183 (8.85 %) patients were diagnosed with the disease. In a primary univariate analysis, patients with COVID‐19 were younger, more likely to be male, had a higher HbA1c and BMI, belonged to a lower SES and had fewer pre‐existing medical conditions (Table 1). Furthermore, among COVID‐19 patients, more patients had an HbA1c ≥ 9% (12.02% vs. 7.27%; p < 0.05; crude OR of 1.84; 95% CI 1.13–2.99; Table 2). These differences may reflect infection ‘clusters’ which were characteristic of the spread of COVID‐19 in Israel—the major outbreak cluster was among the ultra‐orthodox Jewish population which is characterized by a younger mean age (leading to fewer prior co‐morbidities) and a lower SES.
TABLE 1.
Clinical characteristics of patients with diabetes and COVID‐19
| Variable | COVID‐19 n = 183 (8.9%) | Control n = 1885 (91.2%) | p value |
|---|---|---|---|
| Mean age, years (CI) | 61.82 (59.9–63.7) | 65.53 (64.8–66.3) | <0.001 |
| Age (years) n (%) | |||
| 0–20 | 0 (0.0) | 3 (0.2) | <0.001 |
| 20–40 | 7 (3.8) | 120 (6.4) | <0.001 |
| 40–60 | 77 (42.1) | 515 (27.3) | <0.001 |
| 60–80 | 81 (44.3) | 818 (43.4) | <0.05 |
| 80+ | 18 (9.8) | 429 (22.8) | <0.001 |
| Low‐medium SES n (%) | 135 (78.5) | 1,131 (62.0) | <0.05 |
| Male n (%) | 109 (59.6) | 934 (49.6) | <0.05 |
| Mean BMI (CI) | 31.4 (30.6–32.2) | 30.1 (29.8–30.5) | <0.05 |
| Mean HbA1c % (CI) | 7.12 (6.81–7.57) | 6.59 (6.52–6.65) | <0.05 |
| Smoking n (%) | 33 (19.3) | 348 (21.4) | 0.52 |
| Depression/anxiety n (%) | 27 (14.8) | 370 (19.6) | 0.055 |
| Schizophrenia n (%) | 8 (4.4) | 57 (3) | 0.15 |
| Dementia n (%) | 14 (7.7) | 261 (13.9) | <0.05 |
| Hypertension n (%) | 98 (53.55) | 1228 (65.15) | <0.005 |
| Ischaemic heart disease n (%) | 39 (21.3) | 648 (34.4) | <0.005 |
| CVA n (%) | 11 (6.0) | 187 (9.9) | 0.0431 |
| CHF n (%) | 16 (8.7) | 257 (13.6) | 0.0620 |
| Chronic lung disease n (%) | 24 (13.1) | 405 (21.5) | 0.0077 |
| Obesity n (%) | 101 (58.1) | 833 (46.9) | 0.0052 |
Abbreviations: BMI, body mass index; CI, confidence interval; CHF, congestive heart failure; CVA, cerebrovascular accident; SES, socio‐economic status.
TABLE 2.
HbA1c and OR for COVID‐19 in patients with diabetes
| Variable | COVID‐19 n = 183 (8.9%) | Control n = 1,885 (91.2%) | Crude OR (95% CI) for COVID‐19 | p value |
|---|---|---|---|---|
| HbA1c categories | ||||
| <7% | 118 (64.5) | 1351 (71.7) | 1.00 | ‐ |
| 7%–7.9% | 24 (13.1) | 258 (13.7) | 1.065 (0.67–1.69) | 0.79 |
| 8%–8.9% | 19 (10.4) | 139 (7.4) | 1.56 (0.93–2.62) | 0.09 |
| ≥9% | 22 (12.0) | 137 (7.3) | 1.84 (1.13–2.99) | < 0.05 |
Abbreviations: CI, confidence interval; OR, odds ratio.
We conducted a complementary analysis among patients who were COVID‐19 negative. Patients who were found negative to COVID‐19 but with an HbA1c ≥ 9% were younger (mean age 59.9 [95% CI 57.5–62.4] vs. 65.9 [95% CI 65.2–66.7]), male (82 [59.9%] vs 852 [48.8%]) and from a lower SES (103 [77.4%] vs. 1028 [60.8%]). No differences were observed in BMI or prevalence of pre‐existing medical conditions (depression, dementia, HTN, CVA, IHD, and chronic lung disease) in these patients.
3.2. Clinical characteristics of diabetic patients hospitalized due to COVID‐19
Of the 183 patients with COVID‐19, 46 were admitted to hospital. The mean HbA1c in hospitalized patients was higher than patients who were not hospitalized. More hospitalized patients had an HbA1c between 8% and 8.9% or ≥9% when compared to patients who were not hospitalized (Tables 3 and 4). Other factors significantly associated with hospitalization were older age, female sex, higher SES, CVA, and chronic lung disease. In our cohort, obesity and smoking status were not associated with an increased risk for hospitalization. In a multivariate analysis, controlling for multiple prior clinical conditions, the only parameter associated with a significantly increased risk for hospitalization was HbA1c ≥ 9% (adjusted OR 4.95; 95% CI 1.55–15.76; p < 0.05). Other variables had no significant impact on the risk for hospitalization due to COVID‐19 (Table 5).
TABLE 3.
Clinical characteristics of patients with diabetes hospitalized due to COVID‐19
| Variable N (%) | Hospitalized n = 46 (25.1%) | Not hospitalized n = 137 (74.9%) | p value |
|---|---|---|---|
| Mean age, years (CI) | 67.0 (63.0–71.1) | 60.0 (58.0–62.2) | ‐ |
| Age (years) n (%) | <0.001 | ||
| 0–20 | 0 (0.0) | 0 (0.0) | ‐ |
| 20–40 | 1 (2.2) | 6 (4.4) | <0.001 |
| 40–60 | 14 (30.4) | 63 (46.0) | <0.01 |
| 60–80 | 21 (45.7) | 60 (43.8) | 0.124 |
| 80+ | 10 (21.7) | 8 (5.8) | <0.05 |
| Low‐medium SES n (%) | 28 (63.6) | 107 (83.6) | 0.05 |
| Male n (%) | 25 (54.3) | 84 (61.3) | 0.05 |
| Mean BMI (CI) | 31.8 (30.2–33.4) | 30.6 (29.7–31.5) | 0.195 |
| Mean HbA1c % (CI) | 7.75 (7.17–8.32) | 6.83 (6.54–7.13) | <0.005 |
| Smoking n (%) | 10 (24.4) | 23 (17.7) | 0.20 |
| Depression/anxiety n (%) | 9 (19.6) | 18 (13.1) | 0.34 |
| Schizophrenia n (%) | 5 (10.9) | 9 (6.6) | 0.40 |
| Dementia n (%) | 1 (2.2) | 7 (5.1) | 0.34 |
| Hypertension n (%) | 28 (60.9) | 70 (51.1) | 0.25 |
| Ischaemic heart disease n (%) | 13 (28.3) | 26 (19.0) | 0.18 |
| CVA n (%) | 4 (8.7) | 7 (5.1) | <0.05 |
| CHF n (%) | 8 (17.4) | 8 (5.8) | 0.38 |
| Chronic lung disease n (%) | 13 (28.3) | 11 (8.0) | <0.001 |
| Obesity n (%) | 25 (56.8) | 76 (58.5) | 0.85 |
Abbreviations: BMI, body mass index; CI, confidence interval; CHF, congestive heart failure; CVA, cerebrovascular accident; SES, socio‐economic status.
TABLE 4.
HbA1c and OR for hospitalization in patients with diabetes and COVID‐19
| Variable | Hospitalized n = 46 (25.1%) | Not hospitalized n = 137 (74.9%) | Crude OR (95% CI) for being hospitalized | p value |
|---|---|---|---|---|
| Haemoglobin HbA1c categories | ||||
| <7% | 20 (43.5) | 95 (69.3) | 1.00 | ‐ |
| 7%–7.9% | 8 (17.4) | 16 (11.7) | 2.38 (0.88–6.39) | 0.08 |
| 8%–8.9% | 8 (17.4) | 12 (8.8) | 3.17 (1.11–8.97) | <0.05 |
| ≥9% | 10 (21.7) | 14 (10.2) | 3.391 (1.28–9.28) | <0.01 |
Abbreviations: CI, confidence interval; OR, odds ratio.
TABLE 5.
Multivariate logistic regression analysis, controlling for multiple clinical conditions assessing the OR for hospitalization in patients with diabetes and COVID‐19
| Variable | Adjusted OR (95% CI)* | p value |
|---|---|---|
| HbA1c ≥ 9 | 4.95 (1.55–15.76) | <0.01 |
| Age | 1.05 (0.99–1.10) | 0.09 |
| Male sex | 0.70 (0.31–2.07) | 0.64 |
| Low‐medium SES | 0.53 (0.12–1.00) | 0.06 |
| Smoking | 2.29 (0.73–7.19) | 0.15 |
| Depression/anxiety | 2.07 (0.63–6.79) | 0.23 |
| Schizophrenia | 0.41 (0.03–5.19) | 0.49 |
| Dementia | 0.16 (0.01–1.34) | 0.09 |
| Hypertension | 0.71 (0.24–2.06) | 0.53 |
| Ischaemic heart disease | 0.43 (0.20–2.34) | 0.55 |
| CVA | 0.73 (0.13–3.89) | 0.71 |
| CHF | 5.41 (0.99–27.36) | 0.05 |
| Chronic lung disease | 2.6 (0.77–8.76) | 0.12 |
| Obesity | 0.96 (0.39–2.45) | 0.98 |
Abbreviations: BMI, body mass index; CI, confidence interval; CHF, congestive heart failure; CVA, cerebrovascular accident; OR, odds ratio; SES, socio‐economic status.
*Adjusted for age, sex, SES and co‐morbidities.
We repeated the multivariate analysis with an HbA1c cutoff of >7% or with HbA1c as a continuous variable. In the former analysis, the OR for hospitalization in patients with an HbA1c > 7% was 6.07 (95%CI 2.36–15.62; p < 0.005). Furthermore, an increased risk for hospitalization was observed in patients with prior congestive heart failure and a reduced risk for hospitalization in patients with low SES (Table S1). In the latter analysis, any increase in HbA1c by 1% above a 5% baseline was associated with an OR for hospitalization of 1.46 (95% CI 1.14–1.85; p = 0.002). In this model, low SES was protective and age was a risk factor for hospitalization (Table S2).
4. DISCUSSION
Using pre‐infection glycaemic control data, we identify HbA1c as a clear predictor of COVID‐19 severity (as assessed by the need for hospitalization). Other clinical characteristics which were significantly linked to hospitalization included female gender, low SES, a prior CVA, and chronic lung disease. Surprisingly, obesity and smoking status and male sex were not associated with hospitalization despite being suggested as risk factors in prior publications. 13 , 14 , 15 In a multivariate logistic regression model adjusting for multiple potential risk factors and chronic conditions which may have a deleterious effect on disease outcomes (including age, sex, smoking, IHD, SES, depression/anxiety, schizophrenia, dementia, hypertension, CVA, CHF, chronic lung disease, and obesity), only HbA1c ≥ 9% remained a significant predictor for hospitalization. In fact, HbA1c remained a strong predictor of hospitalization due to COVID‐19 when the model was repeated with an HbA1c cutoff of >7% or when addressing HbA1c as a continuous variable.
Prior studies linking glycaemic control and COVID‐19 have suggested an association between in hospital glucose levels and disease severity. The link between hyperglycaemia during hospitalization and disease severity has been previously reported in numerous studies covering multiple disease conditions. 16 , 17 , 18 However, tight glycaemic control failed to lead to improved outcomes in hospitalized patients in conditions other than COVID‐19. 19 This may suggest that hyperglycaemia is a biomarker for the severity of the disease and poor overall health status rather than having an actual causative effect. To our knowledge, our study is the first to identify pre‐infection glycaemic control as a risk factor for COVID‐19 severity.
The weaknesses of our study are its relatively small number of hospitalized patients, retrospective design, and lack of information regarding disease severity beyond the need for hospitalization. It would be interesting to learn whether HbA1c predicts COVID‐19 mortality, need for ICU stay, or need for artificial ventilation. Moreover, our study does not prove that pre‐infection glycaemic control has a causative role in COVID‐19 severity. Similar to hyperglycaemia during hospitalization, HbA1c may be a biomarker of poor health. Future mechanistic studies are needed to determine this relationship. However, the multivariate logistic regression statistical model identifying HbA1c as the only predictor of COVID‐19 hospitalization among numerous other severe and chronic health conditions (including prior cardiovascular disease, mental illness, chronic lung disease, etc.) and the observation that risk for hospitalization increases already at HbA1c ≥ 8% may suggest its pathogenic importance and clinical utility in risk stratification. Other strengths of our study are the large and comprehensive cohort (including all COVID‐19‐tested patients with diabetes in a large HMO) and the large amount of pre‐COVID‐19 clinical background information which we could access and present.
As many countries continue to battle with the COVID‐19 pandemic while others begin to release strict measures of social distancing, identifying high‐risk populations is key to successfully overcome this disease. Paying special attention to patients with diabetes and an HbA1c ≥ 9 while allowing a more lenient approach to patients with well controlled disease may prove to be beneficial in minimizing the economic and social burden associated with an undiscriminating approach. Moreover, once a vaccine is available, prioritizing those who are at highest risk for severe COVID‐19 will be crucial to efficiently overcome this disease.
CONFLICT OF INTEREST
All the authors declared that they have no conflict of interest.
AUTHOR CONTRIBUTIONS
Eugene Merzon, Roy Eldor, Ilan Green, Avivit Golan‐Cohen, Miriam Shpigelman, Shlomo Vinker, and Itamar Raz contributed the research question; Eugene Merzon and Ilan Green performed data mining; and Eugene Merzon performed statistical analysis. Roy Eldor, Eugene Merzon, Ilan Green, and Miriam Shpigelman wrote the final draft of the manuscript. Shlomo Vinker, Itamar Raz, and Avivit Golan‐Cohen contributed to editing of the manuscript. No honorarium, grant, or other form of payment was given to any of the authors to produce the manuscript.
ETHICAL STATEMENT
This is a data‐based study, and as such, has no clinical trial registration number. The study received IRB approval from the Shamir Medical Centre IRB.
Supporting information
Supplementary Material
References
REFERENCES
- 1. Huang I, Lim MA, Pranata R. Diabetes mellitus is associated with increased mortality and severity of disease in COVID‐19 pneumonia ‐ a systematic review, meta‐analysis, and meta‐regression. Diabetes Metab Syndr. 2020;14(4):395‐403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Roncon L, Zuin M, Rigatelli G, Zuliani G. Diabetic patients with COVID‐19 infection are at higher risk of ICU admission and poor short‐term outcome. J Clin Virol. 2020;127:104354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Guo W, Li M, Dong Y, et al. Diabetes is a risk factor for the progression and prognosis of COVID‐19. Diabetes Metab Res Rev. 2020:e3319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Yehya A, Carbone S. Managing type 2 diabetes mellitus during COVID‐19 pandemic: the bittersweet. Diabetes Metab Res Rev. 2020:e3360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Wang F, Yang Y, Dong K, et al. Clinical characteristics of 28 patients with diabetes and Covid‐19 in Wuhan, China. Endocr Pract. 2020;26(6):668–674. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Wu C, Chen X, Cai Y, et al. Risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease 2019 pneumonia in Wuhan, China. JAMA Intern Med. 2020;180(7):934–943. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Li X, Xu S, Yu M, et al. Risk factors for severity and mortality in adult COVID‐19 inpatients in Wuhan. J Allergy Clin Immunol. 2020;146(1):110–118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Iacobellis G, Penaherrera CA, Bermudez LE, Bernal Mizrachi E. Admission hyperglycemia and radiological findings of SARS‐COv2 in patients with and without diabetes. Diabetes Res Clin Pract. 2020;164:108185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Zhu L, She Z, Cheng X, et al. Association of blood glucose control and outcomes in patients with COVID‐19 and pre‐existing type 2 diabetes. Cell Metab. 2020;31(6):1068–1077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Jaffe DH, Shmueli A, Ben‐Yehuda A, et al. Community healthcare in Israel: quality indicators 2007‐2009. Isr J Health Pol Res. 2012;1(1):3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Tsibel N, Badran Y, Burck L, et al. Characterization and classification of geographical units by the socio‐economic level of the population . 2015. https://www.cbs.gov.il/en/publications/Pages/2019/Characterization‐and‐Classification‐of‐Geographical‐Units‐by‐the‐Socio‐Economic‐Level‐of‐the‐Population‐2015.aspx. Accessed September 18, 2020.
- 12. Rennert G, Peterburg Y. Prevalence of selected chronic diseases in Israel. Isr Med Assoc J. 2001;3(6):404‐408. [PubMed] [Google Scholar]
- 13. Garg S, Kim L, Whitaker M, et al. Hospitalization rates and characteristics of patients hospitalized with laboratory‐confirmed coronavirus disease 2019 ‐ COVID‐NET, 14 states, March 1‐30, 2020. MMWR Morb Mortal Wkly Rep. 2020;69(15):458‐464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Simonnet A, Chetboun M, Poissy J, et al. High prevalence of obesity in severe acute respiratory syndrome coronavirus‐2 (SARS‐CoV‐2) requiring invasive mechanical ventilation. Obesity. 2020;28(7):1195–1199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Zheng Z, Peng F, Xu B, et al. Risk factors of critical & mortal COVID‐19 cases: a systematic literature review and meta‐analysis. J Infect. 2020;81(2):e16–e25. 10.1016/j.jinf.2020.04.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, Kitabchi AE. Hyperglycemia: an independent marker of in‐hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab. 2002;87(3):978‐982. [DOI] [PubMed] [Google Scholar]
- 17. Capes SE, Hunt D, Malmberg K, Gerstein HC. Stress hyperglycaemia and increased risk of death after myocardial infarction in patients with and without diabetes: a systematic overview. Lancet. 2000;355(9206):773‐778. [DOI] [PubMed] [Google Scholar]
- 18. Williams LS, Rotich J, Qi R, et al. Effects of admission hyperglycemia on mortality and costs in acute ischemic stroke. Neurology. 2002;59(1):67‐71. [DOI] [PubMed] [Google Scholar]
- 19. Wiener RS, Wiener DC, Larson RJ. Benefits and risks of tight glucose control in critically ill adults: a meta‐analysis. J Am Med Assoc. 2008;300(8):933‐944. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Material