Abstract
Objectives
Patients with diabetes are potentially at higher risk of mortality due to coronavirus disease-2019 (COVID-19). In this study, we aimed to compare the outcomes and severity of pulmonary involvement in COVID-19 patients with and without diabetes.
Methods
In this cohort study, we recruited patients with diabetes who were hospitalized due to COVID-19 during the period from February 2020 to May 2020. Hospitalized individuals without diabetes were enrolled as control subjects. All patients were followed for 90 days and clinical findings and patients’ outcomes were reported.
Results
Over a period of 4 months, 127 patients with diabetes and 127 individuals without diabetes with a diagnosis of COVID-19 were recruited. Their mean age was 65.70±12.51 years. Mortality was higher in the group with diabetes (22.8% vs 15.0%; p=0.109), although not significantly. More severe pulmonary involvement (p=0.015), extended hospital stay (p<0.001) and greater need for invasive ventilation (p=0.029) were reported in this population. Stepwise logistic regression revealed that diabetes was not independently associated with mortality (p=0.092). Older age (odds ratio [OR], 1.054; p=0.003), aggravated pulmonary involvement on admission (OR, 1.149; p=0.001), presence of comorbidities (OR, 1.290; p=0.020) and hypothyroidism (OR, 6.576; p=0.021) were associated with mortality. Diabetic foot infection had a strong positive correlation with mortality (OR, 49.819; p=0.016), whereas insulin therapy had a negative correlation (OR, 0.242; p=0.045).
Conclusions
The mortality rate due to COVID-19 did not differ significantly between patients with or without diabetes. Older age, macrovascular complications and presence of comorbidities could increase mortality in people with diabetes. Insulin therapy during hospitalization could attenuate the detrimental effects of hyperglycemia and improve prognosis of patients with COVID-19 and diabetes.
Keywords: cohort study, comorbidity, COVID-19, death, diabetes mellitus, prognosis
Résumé
Objectifs
Les patients diabétiques sont potentiellement exposés à un risque plus élevé de mortalité liée à la maladie à coronavirus 2019 (COVID-19). Dans la présente étude, nous avions pour objet de comparer les résultats cliniques et la gravité de l’atteinte pulmonaire chez les patients diabétiques ou non diabétiques atteints de la COVID-19.
Méthodes
Dans cette étude de cohorte, nous avons recruté les patients diabétiques qui étaient hospitalisés en raison de la COVID-19 au cours de la période de février 2020 à mai 2020. Les sujets témoins recrutés étaient les individus non diabétiques hospitalisés. Nous avons rapporté les résultats cliniques et l’évolution de l’état de santé de tous les patients suivis durant 90 jours.
Résultats
Au cours de la période de 4 mois, nous avons recruté 127 patients diabétiques et 127 individus non diabétiques qui recevaient un diagnostic de COVID-19. L’âge moyen était de 65,70 ± 12,51 ans. La mortalité était plus élevée dans le groupe des patients diabétiques (22,8 % vs 15,0 %; p = 0,109), quoique de façon non significative. Dans cette population, nous avons rapporté l’atteinte pulmonaire plus grave (p = 0,015), le séjour prolongé à l’hôpital (p < 0,001) et la nécessité d’une ventilation effractive (p = 0,029). La régression logistique séquentielle a révélé que le diabète n’était pas indépendamment associé à la mortalité (p = 0,092). Le grand âge (ratio d’incidence approché [RIA], 1,054; p = 0,003), l’atteinte pulmonaire plus grave à l’admission (RIA, 1,149; p = 0,001), la comorbidité (RIA, 1,290; p = 0,020) et l’hypothyroïdisme (RIA, 6,576; p=0,021) étaient associés à la mortalité. Il existait une forte corrélation positive entre l’infection du pied diabétique et la mortalité (RIA, 49,819; p = 0,016), mais une corrélation négative avec l’insulinothérapie (RIA, 0,242; p = 0,045).
Conclusions
Le taux de mortalité lié à la COVID-19 ne diffère pas de façon significative entre les patients diabétiques et les patients non diabétiques. Le grand âge, les complications macrovasculaires et la comorbidité pourraient faire augmenter la mortalité chez les personnes diabétiques. L’insulinothérapie en cours d’hospitalisation pourrait permettre d’atténuer les effets néfastes de l’hyperglycémie et d’améliorer le pronostic des patients diabétiques atteints de la COVID-19.
Mots clés: étude de cohorte, comorbidité, COVID-19, décès, diabète sucré, pronostic
Key Messages.
-
•
Patients with type 2 diabetes macrovascular complications are at higher risk of mortality due to coronavirus disease-2019.
-
•
Blood glucose control and insulin therapy in type 2 diabetes patients diagnosed with severe and critical forms of coronavirus disease-2019 are major factors during hospitalization.
Introduction
Coronavirus disease-2019 (COVID-19), caused by the novel coronavirus known as severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), can lead to severe pneumonia and multiorgan failure, especially in older patients and those with comorbidities, such as diabetes mellitus (DM), hypertension and cardiovascular disorders (1).
Diabetes has deleterious effects on the immune system and could lead to higher susceptibility to bacterial infections (2,3). In SARS-CoV-1 disease, it has been shown that individuals with pre-existing DM are at increased risk of death (4). Also, in previous retrospective studies, DM has been associated with an increased risk of mortality in SARS-CoV-2 infection (5,6). Until now, cohort studies regarding prognosis of patients with coexistent diabetes and COVID-19 with the same age and sex distribution have not been done.
In this cohort study performed on patients hospitalized for COVID-19, we aimed to determine whether patients with diabetes had worse clinical outcomes and more severe radiologic findings when compared to patients without diabetes.
Methods
Hospital-based cohort study
In this cohort study, 127 patients with diabetes hospitalized for COVID-19 and 127 control subjects without diabetes, also admitted due to COVID-19, were included. All participants were recruited from the Imam Hossein Medical Center, affiliated with the Shahid Beheshti University of Medical Sciences, Tehran, Iran, from February 2020 to May 2020. The ethics committee of the Shahid Beheshti University of Medical Sciences approved the study and all patients signed written informed consent before enrolment. Diagnosis of COVID-19 was confirmed for participants in both the case and control groups by reverse transcript‒polymerase chain reaction and/or computed tomography (CT) (7) scan findings. To control blood glucose levels during hospitalization, insulin intensification, basal-bolus insulin regimen, sliding scale or insulin infusion protocols were considered for patients with DM to reach and maintain glucose level between 140 and 180 mg/dL (7.8 to 10.0 mmol/L). Patients were visited daily during hospitalization and followed up for 90 days. All patients received COVID-19 treatment based on the latest interim national guideline for diagnosis and management of COVID-19 (8).
Chest CT interpretation
Patients underwent chest CT examination after admission, and all CT scans were reviewed and reported by the same experienced radiologist. The severity of pulmonary involvement was reported based on a quantitative scoring system (9). A scoring scale from 0 to 5 was considered for each of the lung lobes, based on visual inspection (0 for no involvement in a particular lobe, and 5 for >75% involvement); a score of 1 was considered for <5%, 2 for 5% to 25%, 3 for 26% to 49% and 4 for 50% to 75% of each lobe being involved. The maximum score of 25 was considered when there was involvement of >75% of all 5 lobes of the lung. The predominant patterns on chest CT imaging were classified into 3 groups: ground-glass opacity (GGO), consolidation and GGO/consolidation (mixed). Secondary CT findings, such as pleural effusion, pericardial effusion, cardiomegaly and lymphadenopathy (10) >10 mm, were also recorded. Distribution of lung lesions was grouped into subpleural, peribronchovascular and perihilar categories.
Data collection
The attending physician collected demographic data for all recruited patients. Primary clinical and laboratory data, including glycemic control profile, inflammatory marker panel, complete blood count with differentiation, renal and hepatic function profiles, electrolytes, blood pressure, ventilation and oxygen saturation status, were documented. All patients were assessed with regard to comorbidities, underlying diseases and drug history. The data for type of hypoglycemic agents (i.e. insulin therapy or oral hypoglycemic agents) given to patients with diabetes were also compiled.
Primary outcome
Mortality rate was assessed during hospitalization and for up to 90 days after disease onset.
Secondary outcomes
Oxygen requirement, mechanical ventilation, duration of hospital stay, acute respiratory distress syndrome (11) occurrence (11,12), shock and multiorgan failure were all considered as secondary outcomes (13). The Charlson index was utilized to assess the prognostic effects of comorbidities (14).
Statistical analysis
Analysis was performed using STATA version 14 (StataCorp, College Station, Texas, United States) and R version 3.6.2 (R Foundation for Statistical Computing, Vienna, Austria). Data were assessed for parametric and nonparametric distribution by the Kolmogorov-Smirnov test. Quantitative variables with a normal distribution are reported as mean ± standard deviation and those with a non-normal distribution are reported as median (interquartile range). Qualitative variables are presented as frequency and percent. Categorical data were analyzed by chi-square test or Fisher’s exact test (if >25% of the categories had frequencies <5). Differences in continuous data were compared using a t test or a Mann-Whitney U test in the bivariant situation. p<0.05 was considered significant. We used multivariable logistic regression to evaluate the association between covariates and outcomes among the patients studied. Glycated hemoglobin (A1C) stepwise selection methods were used (backward and forward) to select a set of candidate predictors for inclusion in the multivariate model. The overall performance of models was evaluated using the Brier score, Nagelkerke’s R2 and area under the curve were calculated for discrimination, and the Hosmer-Lemeshow test was used for evaluation of calibration. Odds ratios and 95% confidence intervals were calculated. p<0.05 was considered significant for uni- and multivariate regression analyses.
The study was approved by the ethics committee Shahid Beheshti University of Medical Sciences (Code No. IR.SBMU.RETECH.REC.1399.063).
Results
By the end of the study period, 127 patients with diabetes (case group) and 127 patients without diabetes (control group) had been enrolled and followed up for 90 days. Participants were 65.70±12.51 years of age (mean ± standard deviation). Baseline demographics on admission to the laboratory and physical findings are presented in Table 1 . All patients in the case group had type 2 diabetes. Ten patients were diagnosed with type 2 diabetes during hospitalization for COVID-19. Twenty-three patients had a history of type 2 diabetes of ≤5 years. Thirty-eight, 24 and 32 patients had a history of type 2 diabetes for 5 to 10 years, 10 to 15 years and >15 years, respectively. No correlation was found between the duration of diabetes and mortality (p=0.317). Mean A1C level for patients with diabetes was 9.15±2.21%. According to the latest guidelines for glycemic control from the American Diabetes Association and Diabetes Canada (15,16), 97 patients (76%) had uncontrolled DM at the time of admission based on their initial A1C level of >7%. DM was treated with insulin in 39 patients. In 78 and 10 patients, oral hypoglycemic agents and nutritional diet intervention, respectively, were utilized. During hospitalization, a basal-bolus regimen, sliding scale and insulin infusion were considered for 55, 10 and 4 patients, respectively, with poorly controlled diabetes. Furthermore, 28 patients received insulin intensification for uncontrolled blood glucose. Eventually, blood glucose levels in 78% of patients with diabetes were controlled during hospitalization. Six patients had at least 1 episode of hypoglycemia (defined as blood glucose levels <70 mg/dL [<3.9 mmol/L]) (17,18).
Table 1.
Patients’ demographics and baseline clinical and laboratory findings
| Characteristics | Total (n=254) | Cases (n=127) | Controls (n=127) | p Value |
|---|---|---|---|---|
| Age, years | 65.70±12.51 | 66.38±12.51 | 65.03±12.53 | 0.332 |
| Sex | ||||
| Male | 142 (55.9) | 71 (55.9) | 71 (55.9) | 1.000 |
| Female | 112 (44.1) | 56 (44.1) | 56 (44.1) | |
| Positive COVID-19 RT-PCR | 194 (76.4) | 103 (81.1) | 91 (71.7) | 0.191 |
| Body mass index, kg/m2 | 26.12 (5.16) | 26.67 (5.83) | 25.71 (4.12) | 0.001 |
| Systolic blood pressure, mmHg | 119.84±16.69 | 120.86±18.37 | 118.84±14.87 | 0.411 |
| Diastolic blood pressure, mmHg | 77.00±9.51 | 77.64±9.69 | 76.36±9.80 | 0.443 |
| CT scan severity index | 10.00 (7.00) | 11.00 (6.00) | 10.00 (6.00) | 0.015 |
| Pulse rate, beats/min | 88.40±14.99 | 88.37±15.44 | 88.42±14.59 | 0.850 |
| Respiratory rate, breaths/min | 18.00 (4.00) | 18.00 (2.00) | 18.00 (4.00) | 0.017 |
| Oxygen saturation | 91.00 (6.00) | 90.00 (8.00) | 93.00 (5.00) | <0.001 |
| ARDS at baseline | 27 (10.6) | 14 (11.0) | 13 (10.0) | 0.841 |
| Laboratory data | ||||
| Serum creatinine, mg/dL | 1.20 (0.60) [0.50–25.00] | 1.30 (0.80) [0.70–25.00] | 1.20 (0.5) [0.50–8.10] | 0.078 |
| Estimated glomerular filtration rate | 0.161 | |||
| >90 mL/min | 17 (7.42) | 5 (4.50) | 12 (9.40) | |
| 60–90 mL/min | 76 (33.19) | 35 (31.50) | 41 (32.30) | |
| 30–60 mL/min | 98 (42.80) | 48 (43.30) | 50 (39.40) | |
| <30 mL/min | 38 (16.59) | 23 (20.7) | 15 (11.80) | |
| C-reactive protein, mg/dL | 53.00 (45.25) [16.00–72.10] | 64.50 (61.88) [2.80–287.00] | 54.60 (65.10) [6.00–239.00] | 0.664 |
| Erythrocyte sedimentation rate, seconds | 47.91±26.10 | 47.92±27.72 | 47.90±24.51 | 0.900 |
| Lactate dehydrogenase, units/L | 483.00 (552.00) [374.00–981.00] | 334.50 (259.00) [84.00–9,359.00] | 427.00 (382.00) [170.00–1,746.00] | 0.168 |
| Aspartate aminotransferase, units/L | 46.00 (24.00) [23.00-51.00] | 31.500 (16.00) [11.00-2127] | 33.00 (26.50) [10.00–1,786.00] | 0.829 |
| Alanine aminotransferase, units/L | 23.00 (18.00) [17.00–38.00] | 27.00 (11.00) [13.00–1,314.00] | 25.00 (25.00) [10.00–912.00] | 0.386 |
| Procalcitonin, ng/mL | 0.23 (0.37) [0.07‒0.49] | 0.30 (0.36) [0.07–8.94] | 0.28 (1.11) [0.01–26.94] | 0.854 |
| White blood cells, cells/μL | 6,100.00 (3,250.00) [4,000.00–9,400.00] | 6,800.00 (4,550.00) [2,900.00–26,400.00] | 6,400.00 (7,550.00) [2,600.00–27,400.00] | 0.066 |
| Neutrophils, % | 72.42±12.04 | 73.44±12.19 | 71.42±11.85 | 0.287 |
| Lymphocyte, % | 20.27±12.03 | 19.93±11.39 | 20.61±10.54 | 0.554 |
| Hemoglobin, g/dL | 12.50±2.00 | 12.32±2.01 | 12.68±1.98 | 0.052 |
| Platelet count ×106, cells/μL | 199.98±86.26 | 208.08±91.67 | 191.40±79.62 | 0.131 |
| D-dimer, units/mL | 564.00 (191.00) [338.00‒640.00] | 513.50 (1,673.00) [76.00‒7,500.00] | 576.00 (1,565.00) [140.00–3,888.00] | 0.851 |
| Albumin, g/dL | 3.80±0.56 | 3.76±0.51 | 3.85±0.61 | 0.052 |
| Creatinine phosphokinase, units/L | 123.00 (156.00) [18.00–2,294.00] | 146.50 (197.00) [35.00–1,387.00] | 113.50 (126.00) [18.00–2,294.00] | 0.050 |
| 25-dihydroxyvitamin D3, ng/mL | 22.75±12.94 | 24.18±14.33 | 20.88±10.74 | 0.393 |
| Smoking history | 0.554 | |||
| Nonsmoker | 240 (94.5) | 119 (93.7) | 121 (95.3) | |
| Ex-smoker | 5 (2.0) | 2 (1.6) | 3 (2.4) | |
| Currently smoking | 9 (3.5) | 6 (4.7) | 3 (2.4) | |
| Past medical history | ||||
| Coexisting disorder | <0.001 | |||
| Positive | 184 (72.4) | 118 (92.9) | 66 (52.0) | |
| Negative | 70 (27.6) | 9 (7.1) | 61 (48.0) | |
| Chronic obstructive pulmonary disease | 7 (2.8) | 4 (3.1) | 3 (2.4) | 0.702 |
| Hypertension | 109 (42.9) | 68 (53.5) | 41 (32.3) | 0.001 |
| Cerebrovascular accident | 21 (8.3) | 16 (12.6) | 5 (3.9) | 0.012 |
| Malignancy | 11 (4.3) | 4 (3.1) | 7 (5.5) | 0.355 |
| Immunodeficiency | 5 (2.0) | 4 (3.1) | 1 (0.8) | 0.175 |
| Chronic kidney disease | 21 (8.3) | 16 (12.6) | 5 (3.9) | 0.012 |
| End-stage renal disease/dialysis | 14 (5.5) | 9 (7.1) | 5 (3.9) | 0.271 |
| Ischemic heart disease | 64 (25.2) | 39 (30.7) | 25 (19.7) | 0.043 |
| Hypothyroidism | 12 (4.7) | 6 (4.7) | 6 (4.7) | 1.000 |
| Diabetic foot infection | 6 (2.4) | 6 (4.7) | 0 (0) | 0.013 |
| On admission signs and symptoms | ||||
| Cough | 163 (64.6) | 89 (70.1) | 74 (58.3) | 0.066 |
| Cough type | 0.341 | |||
| Nonproductive | 129 (79.8) | 73 (82.0) | 56 (75.7) | |
| Productive | 34 (20.2) | 16 (18.0) | 18 (24.3) | |
| Fatigue | 92 (36.2) | 49 (38.6) | 43 (33.9) | 0.433 |
| Nausea/vomiting | 44 (17.3) | 26 (20.5) | 18 (14.2) | 0.185 |
| Abdominal pain | 16 (6.3) | 7 (5.5) | 9 (7.1) | 0.605 |
| Diarrhea | 16 (6.3) | 11 (8.7) | 5 (3.9) | 0.121 |
| Chills | 62 (24.4) | 33 (26.0) | 29 (22.8) | 0.559 |
| Fever | 109 (42.9) | 53 (41.7) | 56 (44.1) | 0.704 |
| Anorexia | 43 (16.9) | 21 (16.5) | 22 (17.3) | 0.867 |
| Headache | 20 (7.9) | 10 (7.9) | 10 (7.9) | 1.000 |
| Loss of consciousness | 16 (6.3) | 11 (8.7) | 5 (3.9) | 0.121 |
| Seizure | 5 (2.0) | 4 (3.1) | 1 (0.8) | 0.175 |
| Myalgia | 74 (29.1) | 36 (28.3) | 38 (29.9) | 0.782 |
| Chest pain | 29 (11.4) | 14 (11.0) | 15 (11.8) | 0.861 |
| Dyspnea | 152 (59.8) | 83 (65.4) | 69 (54.3) | 0.073 |
ARDS, acute respiratory distress syndrome; COVID-2019, coronavirus disease-2019; CT, computed tomography; RT-PCR, reverse transcript‒polymerase chain reaction.
Note: Data expressed as mean ± standard deviation, number (%) or number (%) [range].
No significant differences were observed for frequency of steroid utilization in the 2 groups (p=0.197); steroids were used in 7 (5%) and 3 (2%) individuals in the case and control groups, respectively. Four patients in the case group were hospitalized due to recent cerebrovascular accident, after which they were diagnosed with COVID-19.
Significantly higher respiratory rate (p=0.017) and lower oxygen saturation (p<0.001) were recorded in patients with diabetes on admission; these patients also had more severe pulmonary involvement and higher CT scan severity indices (p=0.0015). At baseline, there was no significant difference in development of acute respiratory distress syndrome (ARDS) between the 2 groups (p=0.841). As shown in Table 2 , GGO was the most common pattern on CT scans of all patients. Compared to patients without diabetes, consolidation was more frequently observed in patients with diabetes (p=0.01). The right and left lower lobes were the lobes with the highest percentage of involvement in all patients (85.4% and 85.8%, respectively). The left upper and lower lobes were more likely to be involved in patients with diabetes compared to their counterparts without diabetes (p=0.07 and p=0.06, respectively). Lung lesions were mostly distributed in the subpleural area (68.1%), a distribution type that was significantly more common among patients without diabetes (p=0.01).
Table 2.
Computed tomography findings of patients
| Variables | Total (n=254) | Cases (n=127) | Controls (n=127) | p Value |
|---|---|---|---|---|
| Pattern of involvement | ||||
| GGO | 139 | 68 | 71 | 0.53 |
| Consolidation | 62 | 40 | 22 | 0.01 |
| GGO/consolidation | 28 | 14 | 14 | 0.93 |
| Distribution of lesions | ||||
| Subpleural | 173 | 79 | 94 | 0.01 |
| Perihilar | 2 | 0 | 2 | 0.24 |
| Peribronchovascular | 98 | 50 | 48 | 0.96 |
| Other findings | ||||
| Cardiomegaly | 3 | 1 | 2 | 0.62 |
| Lymphadenopathy >10 mm | --- | --- | --- | --- |
| Pericardial effusion | 3 | 2 | 1 | 1.00 |
| Pleural effusion | 21 | 9 | 12 | 0.44 |
| Lung lobe involvement | ||||
| Right upper lobe | 205 (80.7) | 108 (85.0) | 97 (76.4) | 0.17 |
| Right middle lobe | 195 (76.8) | 104 (81.9) | 91 (71.7) | 0.11 |
| Right lower lobe | 217 (85.4) | 114 (89.8) | 103 (81.1) | 0.11 |
| Left upper lobe | 209 (82.3) | 111 (87.4) | 98 (77.2) | 0.07 |
| Left lower lobe | 218 (85.8) | 115 (90.6) | 103 (81.1) | 0.06 |
GGO, ground-glass opacity.
Note: Data expressed as number or number (%).
After the 90-day period of follow up, of 254 subjects, 48 (18.9%) had died during the treatment or follow-up period. Ten patients were rehospitalized after discharge. Mortality was higher amongst the population with DM compared with the control group (22.8% to 15.0%), but the difference was not statistically significant (p=0.109). Furthermore, a statistically significant longer hospital stay was recorded in patients with diabetes (7 vs 5 days, p<0.001). ARDS, septic shock and multiorgan failure were observed in 40, 21 and 25 patients, respectively, during hospitalization. No statistically significant differences were observed between the 2 groups of patients with regard to secondary outcomes (Table 3 ).
Table 3.
Length of hospital stay and outcome
| Variables | Total (n=254) | Cases (n=127) | Controls (n=127) | p Value |
|---|---|---|---|---|
| Ventilation type, n (%) | 0.029 | |||
| Invasive | 27 (10.6) | 19 (15.0) | 8 (6.3) | |
| Noninvasive | 227 (89.4) | 108 (85.0) | 119 (93.7) | |
| Secondary outcomes, % | ||||
| ARDS | 40 | 21 | 19 | 0.432 |
| Septic shock | 21 | 13 | 8 | 0.181 |
| Multiorgan failure | 25 | 15 | 10 | 0.200 |
| Hospital length of stay (days) [range] | 6 (7) [1–60] | 7 (7) [1–60] | 5 (7) [1–34] | <0.001 |
| Outcome, n (%) | 0.109 | |||
| Death | 48 (18.9) | 29 (22.8) | 19 (15.0) | |
| Recovery | 206 (81.1) | 98 (77.2) | 108 (85.0) |
ARDS, acute respiratory distress syndrome.
Logistic regression was performed in both the total population (254 patients) and in the patients with diabetes (n=127). Discrimination indexes, R2=0.433, Brier = 0.102 for total population and R2=0.443, Brier=0.113 in the population with diabetes were calculated. Area under the curve was 0.874 and 0.844 for the model for the total population and DM population, respectively. The goodness-of-fit test (Hosmer-Lemeshow) resulted in p=0.293 and p=0.179 for the logistic regression model in the total population and DM population, respectively. In the multivariate logistic regression analysis for the total population, older age (odds ratio [OR], 1.054 [1.017 to 1.092]; p=0.003), higher CT severity index (OR, 1.149 [1.054 to 1.252]; p=0.001), Charlson index (OR, 1.290 [1.040 to 1.601]; p=0.020) and history of hypothyroidism (OR, 6.576 [1.325 to 32.626]; p=0.021) were associated with mortality. Also, higher oxygen saturation at baseline (OR, 0.939 [0.909 to 0.970]; p<0.001) was shown to be a protective factor. Diabetes was not significantly associated with mortality (p=0.092). Among patients with diabetes, age (OR, 1.065 [1.009 to 1.123]; p=0.023), CT severity index (OR, 1.202 [1.058 to 1.365]; p=0.005), Charlson index (OR, 1.480 [1.106 to 1.981]; p=0.008) and presence of diabetic foot infection (OR, 49.819 [2.0492 to 1211.186]; p=0.016) were associated with higher mortality, but insulin therapy was shown to be a protective factor (OR, 0.242 [0.061 to 0.967]; p=0.045). Higher body mass index (BMI) (p=0.118), A1C level (p=0.585) and controlled vs uncontrolled diabetes based on A1C level (p=0.201) showed no association with mortality in patients with DM.
Discussion
Our cohort study has demonstrated that diabetes per se is not associated with a statistically significant increase in mortality due to COVID-19. However, patients with diabetes had more severe pneumonia and a more severe course of illness; these patients had a lower oxygen saturation and a higher respiratory rate on admission compared with the control group. We observed a significantly greater extent of pulmonary involvement in patients with diabetes by evaluating and scoring the chest CT scans in our population, with the involvement of subpleural area mostly in the lower lobes. This group of patients also had higher rates of intubation, a need for mechanical ventilation support and extended hospitalization.
Our study differs significantly from previous studies, which considered diabetes and uncontrolled blood glucose level as independent risk factors for mortality in COVID-19 (19). In those retrospective studies, patients with diabetes had different age distributions, and the patients were not age- and sex-matched with a control group of patients without diabetes. Also, the sample sizes in the 2 groups were considerably different (5,6,19,20). In 2 large-scale retrospective population-based studies conducted in England, patients with diabetes were shown to be at increased risk for mortality due to COVID-19 (21,22). Subgroup analysis of the A1C levels in those studied demonstrated that patients with type 2 uncontrolled diabetes are at increased risk of death. In our study, we included individuals with or without diabetes in a 1:1 ratio, with no differences in age and sex distributions as remarkable confounding factors in the mortality of patients with COVID-19.
A notable finding in our study is that duration of diabetes and previous long-term hyperglycemia (evaluated by A1C measurement at admission) did not show any association with mortality. Recently, in the CORONADO study, Cariou et al reported that in patients with coexistent COVID-19 and diabetes, A1C level, representative of long-term glucose control, was not associated with tracheal intubation and/or death, a finding similar to that of our study (7). It has previously been reported that there are no significant differences between the severity of pulmonary involvement and mortality in patients with well-controlled and poorly controlled diabetes and COVID-19 based on A1C levels (23). In individuals with diabetes, acute hyperglycemia during the hospital stay due to stressful conditions, such as hypoxia, fever, medication side effects, cytokine storm and disease severity, could cause insulin resistance (24,25). Furthermore, it has been speculated that SARS coronavirus, via binding angiotensin-converting enzyme-2 (ACE-2) to the pancreas, can destroy beta cells and induce acute hyperglycemia (26). All these factors could weaken the value of A1C as a good prognostication index in patients with COVID-19.
Zhu et al demonstrated that, even though type 2 diabetes correlated with a higher mortality rate in patients with COVID-19, well-controlled blood glucose (upper limit <180 mg/dL per 10 mmol/L) during hospitalization was associated with significantly lower mortality compared to individuals with uncontrolled blood glucose (20). In our study, insulin therapy was considered for all patients in the case group with uncontrolled blood glucose levels to achieve a blood glucose level of 140 to 180 mg/dL (7.8 to 10.0 mmol/L), and the statistical analysis showed a significant negative association between insulin therapy and COVID-19 mortality. Initially, it was hypothesized that patients receiving insulin would be at higher risk of COVID-19 mortality due to uncontrolled and more likely longer durations of diabetes. However, in the subgroup analysis, we observed that patients who received insulin therapy had a decreased mortality risk of about 75%, a finding that could be attributed to optimal blood glucose control during hospitalization and is in line with a previously cited study (20). We do not have data on treatment strategies from studies concluding that patients with uncontrolled diabetes based on A1C levels have higher mortality rates compared to those with controlled diabetes (21,22). It is expected that patients with uncontrolled diabetes would have more frequent episodes of hyperglycemia during their hospital stay, but it is not clear what glucose control strategies were implemented and how effectively these strategies were carried out. In a retrospective study conducted in the United States in 88 hospitals, among 1,122 patients, individuals with uncontrolled hyperglycemia (defined as ≥2 blood glucose test levels at >180 mg/dL within a 24-hour period during hospitalization) were shown to have noticeably higher mortality rates (27).
In our experience, insulin therapy and good control of blood glucose levels could ameliorate the negative effects of hyperglycemia on patient outcomes.
Another factor could be the effects of insulin on the immune system with some immunomodulatory characteristics (28). Given the hyperinflammatory state and cytokine storm occurring during SARS-CoV-2 infection, these characteristics may play a major role, and insulin administration in patients with diabetes and COVID-19 may attenuate the magnitude of the inflammatory response (29).
Obesity is another major factor in the setting of severe infections, due to an impaired immune system, and could also reveal a restrictive pattern in pulmonary function studies (30,31). The CORONADO study showed that BMI was associated with tracheal intubation rate rather than mortality; however, the association was less prominent in patients with morbid obesity (7). Simmonet et al revealed an association between obesity and mechanical ventilation requirement (32), yet our study did not show a significant association between BMI and mortality. It needs to be mentioned that, in our study, the median BMI in the case group was 26.7 kg/m2 compared with 28.4 kg/m2 in the CORONADO study and 29.6 kg/m2 in the study by Simonnet et al (7,32). The difference between the findings of our study and the other 2 studies could be due to ethnical and geographic variations.
Ischemic heart disease, chronic kidney disease, diabetic foot infection and cerebrovascular accident were more prevalent in our patients with diabetes. An unexpected vascular event, such as ischemic stroke, occurred in 4 patients in the DM population, but none in the control group. Controlled diabetes without vascular complications or even short-term, uncomplicated, uncontrolled diabetes were not found to be associated with increased mortality due to COVID-19. Patients with confirmed microvascular complications, such as diabetic foot infection, which occur in patients with long-lasting uncontrolled diabetes (33), are at significantly higher risk of death (7,34). In our subgroup analysis, in individuals with diabetes, there was a significant association between diabetic foot infection and mortality, with a 50-fold increase in the rate of mortality. Low-grade inflammation induced by DM causes damage to the vascular system, and diabetes is a major risk factor for cardiovascular events (35). Death due to cardiovascular events has been shown to be more prevalent in COVID-19 patients (36). However, it is important to consider that DM per se, without vascular complications, may not be associated with increased mortality in COVID-19.
Another major aspect that needs consideration is presence of comorbidities in patients with or without diabetes. Different underlying conditions, such as hypertension, have previously been considered risk factors for severity of COVID-19 (7,37), and, when it coexists with diabetes, more detrimental effects can be expected. In our study, comorbidities were associated with a significant increase in mortality in the total population, and the risk was even higher in patients with DM. We assessed the effect of underlying conditions via the Charlson comorbidity index; a higher Charlson score, that is, a higher number of serious underlying diseases, was associated with an increased risk of mortality.
Interestingly, hypothyroidism was shown to be an independent risk factor of mortality in the total population and could increase mortality by 7-fold in COVID-19 patients without hypothyroidism. It has been demonstrated that hypothyroidism could be related to the decrease in the ACE serum level, and, in turn, this could lead to an increase in ACE receptor expression, which plays a fundamental role in the SARS-CoV-2 cell entry mechanism (38, 39, 40). Thyroid dysfunction could be more prevalent in critically ill patients (41), which may increase mortality in these patients (42,43). Given the critical condition of many COVID-19 patients and the potential effects of hypothyroidism on mortality, it seems reasonable to assess thyroid function in all COVID-19 patients.
Our investigation was done at a single centre, and it lacked racial diversity, which is a study limitation even though the hospital was a referral centre for COVID-19.
Diabetes per se was not associated with significantly higher mortality due to COVID-19. Regardless of baseline A1C levels, insulin therapy and tight control of blood glucose levels during hospitalization can improve the prognosis and decrease the mortality rate of individuals with coexistent COVID-19 and type 2 diabetes. These findings suggest that the greater pulmonary involvement seen on CT scans of patients with diabetes does not necessarily indicate higher mortality likelihood, and thus physicians should be aware that appropriate blood glucose management of such patients with insulin therapy may be beneficial. Our findings show that a history of hypothyroidism increases the risk of mortality. More studies with larger sample sizes are needed to assess the effects of metabolic and endocrine conditions on patients with COVID-19.
Acknowledgment
The authors thank Niloofar Shiva for critical editing of English grammar and syntax of the manuscript.
Disclosure statement
Conflicts of interest: None.
Author Contributions
S.K. conceptualized the study and researched the data, O.M. wrote the manuscript, M.S. reviewed/edited the manuscript, R.M. researched the data, A.K. researched the data and reviewed the manuscript and M.S. researched the data and reviewed/edited the manuscript.
References
- 1.Jordan R.E., Adab P., Cheng K.K. Covid-19: Risk factors for severe disease and death. BMJ. 2020;368:m1198. doi: 10.1136/bmj.m1198. [DOI] [PubMed] [Google Scholar]
- 2.Berbudi A., Rahmadika N., Tjahjadi A.I., Ruslami R. Type 2 diabetes and its impact on the immune system. Curr Diabetes Rev. 2020;16:442–449. doi: 10.2174/1573399815666191024085838. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Knapp S. Diabetes and infection: Is there a link?—A mini-review. Gerontology. 2013;59:99–104. doi: 10.1159/000345107. [DOI] [PubMed] [Google Scholar]
- 4.Yang J.K., Feng Y., Yuan M.Y., et al. Plasma glucose levels and diabetes are independent predictors for mortality and morbidity in patients with SARS. Diabet Med. 2006;23:623–628. doi: 10.1111/j.1464-5491.2006.01861.x. [DOI] [PubMed] [Google Scholar]
- 5.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: 10.1002/dmrr.3319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Targher G., Mantovani A., Wang X.B., et al. Patients with diabetes are at higher risk for severe illness from COVID-19. Diabetes Metab. 2020 doi: 10.1016/j.diabet.2020.05.001. (In press) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Cariou B., Hadjadj S., Wargny M., et al. Phenotypic characteristics and prognosis of inpatients with COVID-19 and diabetes: The CORONADO study. Diabetologia. 2020;63:1500–1515. doi: 10.1007/s00125-020-05180-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Abtahi S.E., Ashrafi F., Iraj B., et al. https://corona.ir/article/flowchart-treatment-covid19-7th-gen Flowchart for Diagnosis and Treatment of Covid 19 Disease in Outpatient and Inpatient Service Levels with Triad Guide for Covid 19 Disease - Seventh Edition.
- 9.Chang Y.C., Yu C.J., Chang S.C., et al. Pulmonary sequelae in convalescent patients after severe acute respiratory syndrome: Evaluation with thin-section CT. Radiology. 2005;236:1067–1075. doi: 10.1148/radiol.2363040958. [DOI] [PubMed] [Google Scholar]
- 10.Laporte M., Naesens L.J. Airway proteases: An emerging drug target for influenza and other respiratory virus infections. Curr Opin Virol. 2017;24:16–24. doi: 10.1016/j.coviro.2017.03.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.ARDS Definition Task Force Acute respiratory distress syndrome: The Berlin definition. JAMA. 2012;307:2526–2533. doi: 10.1001/jama.2012.5669. [DOI] [PubMed] [Google Scholar]
- 12.Brown S.M., Duggal A., Hou P.C., et al. Nonlinear imputation of PaO2/FIO2 from SpO2/FIO2 among mechanically ventilated patients in the ICU: A prospective, observational study. Crit Care Med. 2017;45:1317–1324. doi: 10.1097/CCM.0000000000002514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Seymour C.W., Liu V.X., Iwashyna T.J., et al. Assessment of clinical criteria for sepsis: For the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) JAMA. 2016;315:762–774. doi: 10.1001/jama.2016.0288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Charlson M.E., Pompei P., Ales K.L., MacKenzie C.R. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis. 1987;40:373–383. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 15.American Diabetes Association Glycemic targets: Standards of medical care in diabetes---2020. Diabetes Care. 2020;43(Suppl 1):S66. doi: 10.2337/dc20-S006. [DOI] [PubMed] [Google Scholar]
- 16.Imran S.A., Agarwal G., Bajaj H.S., Ross S. Targets for glycemic control. Can J Diabetes. 2018;42(Suppl. 1):S42–S46. doi: 10.1016/j.jcjd.2017.10.030. [DOI] [PubMed] [Google Scholar]
- 17.American Diabetes Association Glycemic targets: Standards of medical care in diabetes---2019. Diabetes Care. 2019;42(Suppl. 1):S61. doi: 10.2337/dc19-S006. [DOI] [PubMed] [Google Scholar]
- 18.American Diabetes Association Diabetes care in the hospital. Diabetes Care. 2016;39(Suppl. 1):S99. doi: 10.2337/dc16-S016. [DOI] [PubMed] [Google Scholar]
- 19.Huang I., Lim M.A., 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;14:395–403. doi: 10.1016/j.dsx.2020.04.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Zhu L., She Z.G., 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;31:68–77.e3. doi: 10.1016/j.cmet.2020.04.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Holman N., Knighton P., Kar P., et al. Risk factors for COVID-19-related mortality in people with type 1 and type 2 diabetes in England: A population-based cohort study. Lancet Diabetes Endocrinol. 2020;8:823–833. doi: 10.1016/S2213-8587(20)30271-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Barron E., Bakhai C., Kar P., et al. Associations of type 1 and type 2 diabetes with COVID-19-related mortality in England: A whole-population study. Lancet Diabetes Endocrinol. 2020;8:813–822. doi: 10.1016/S2213-8587(20)30272-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Raoufi M., Khalili S., Mansouri M., Mahdavi A., Khalili N. Well-controlled vs poorly-controlled diabetes in patients with COVID-19: Are there any differences in outcomes and imaging findings? Diab Res Clin Pract. 2020;166 doi: 10.1016/j.diabres.2020.108286. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kotas M.E., Medzhitov R.J.C. Homeostasis, inflammation, and disease susceptibility. Cell. 2015;160:816–827. doi: 10.1016/j.cell.2015.02.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Šestan M., Marinović S., Kavazović I., et al. Virus-induced interferon-γ causes insulin resistance in skeletal muscle and derails glycemic control in obesity. Immunity. 2018;49:164–177.e6. doi: 10.1016/j.immuni.2018.05.005. [DOI] [PubMed] [Google Scholar]
- 26.Yang J.-K., Lin S.-S., Ji X.-J., Guo L.-M. Binding of SARS coronavirus to its receptor damages islets and causes acute diabetes. J Acta Diabetol. 2010;47:193–199. doi: 10.1007/s00592-009-0109-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Bode B., Garrett V., Messler J., et al. Glycemic characteristics and clinical outcomes of COVID-19 patients hospitalized in the United States. J Diabetes Sci Technol. 2020;14:813–821. doi: 10.1177/1932296820924469. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.van Niekerk G., Christowitz C., Conradie D., Engelbrecht A.M. Insulin as an immunomodulatory hormone. Cytokine Growth Factor Rev. 2020;52:34–44. doi: 10.1016/j.cytogfr.2019.11.006. [DOI] [PubMed] [Google Scholar]
- 29.Coperchini F., Chiovato L., Croce L., Magri F., Rotondi M. The cytokine storm in COVID-19: An overview of the involvement of the chemokine/chemokine-receptor system. Cytokine Growth Factor Rev. 2020;53:25–32. doi: 10.1016/j.cytogfr.2020.05.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Watson R.A., Pride N.B., Thomas E.L., et al. Reduction of total lung capacity in obese men: Comparison of total intrathoracic and gas volumes. J Appl Physiol (1985) 2010;108:1605–1612. doi: 10.1152/japplphysiol.01267.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Huttunen R., Syrjänen J. Obesity and the risk and outcome of infection. Int J Obes (Lond) 2013;37:333–340. doi: 10.1038/ijo.2012.62. [DOI] [PubMed] [Google Scholar]
- 32.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 (Silver Spring) 2020;28:1195–1199. doi: 10.1002/oby.22831. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Hobizal K.B., Wukich D.K. Diabetic foot infections: current concept review. Diabetic Foot Ankle. 2012;3 doi: 10.3402/dfa.v3i0.18409. 10.3402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Mader J.K., Haas W., Aberer F., et al. Patients with healed diabetic foot ulcer represent a cohort at highest risk for future fatal events. Sci Rep. 2019;9:10325. doi: 10.1038/s41598-019-46961-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Rask-Madsen C., King G.L. Vascular complications of diabetes: Mechanisms of injury and protective factors. Cell Metab. 2013;17:20–33. doi: 10.1016/j.cmet.2012.11.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Zheng Y.Y., Ma Y.T., Zhang J.Y., Xie X. COVID-19 and the cardiovascular system. Nat Rev Cardiol. 2020;17:259–260. doi: 10.1038/s41569-020-0360-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Huang S., Wang J., Liu F., et al. COVID-19 patients with hypertension have more severe disease: A multicenter retrospective observational study. Hypertens Res. 2020;43:824–831. doi: 10.1038/s41440-020-0485-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Vargas F., Rodriguez-Gomez I., Vargas-Tendero P., Jimenez E., Montiel M. The renin-angiotensin system in thyroid disorders and its role in cardiovascular and renal manifestations. J Endocrinol. 2012;213:25–36. doi: 10.1530/JOE-11-0349. [DOI] [PubMed] [Google Scholar]
- 39.Kobori H., Ichihara A., Suzuki H., et al. Role of the renin-angiotensin system in cardiac hypertrophy induced in rats by hyperthyroidism. Am J Physiol. 1997;273:H593–H599. doi: 10.1152/ajpheart.1997.273.2.H593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Marchant C., Brown L., Sernia C. Renin-angiotensin system in thyroid dysfunction in rats. J Cardiovasc Pharmacol. 1993;22:449–455. doi: 10.1097/00005344-199309000-00016. [DOI] [PubMed] [Google Scholar]
- 41.Economidou F., Douka E., Tzanela M., Nanas S., Kotanidou A. Thyroid function during critical illness. Hormones (Athens) 2011;10:117–124. doi: 10.14310/horm.2002.1301. [DOI] [PubMed] [Google Scholar]
- 42.Wang F., Pan W., Wang H., Wang S., Pan S., Ge J. Relationship between thyroid function and ICU mortality: a prospective observation study. Crit Care. 2012;16:R11. doi: 10.1186/cc11151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Montesinos M.D.M., Pellizas C.G. Thyroid hormone action on innate immunity. Front Endocrinol (Lausanne) 2019;10:350. doi: 10.3389/fendo.2019.00350. [DOI] [PMC free article] [PubMed] [Google Scholar]