The burden of COVID-19, caused by the novel SARS-CoV-2, has being increasing worldwide. The World Health Organization declared COVID-19 a pandemic on March 11, 2020 [1]. As of October 6, 2020, a total of 99,733 COVID-19 cases have been reported in the United Arab Emirates with 429 COVID-19–related deaths [2].

Recent studies have suggested that diabetes is a major risk factor contributing to the severity of COVID-19 and resulting mortality [3]. For instance, a meta-analysis of 7 trials in China reported that 9.7% (153/1576) patients with COVID-19 had diabetes [4]. Diabetes is also considered a major risk factor for the development of severe pneumonia and clinical course resulting from COVID-19, and it is reported in approximately 20%-30% of such patients [5]. In addition, poor glycemic control, whether related to diabetes or stress hyperglycemia, is known to be associated with poor patient outcomes, including hospitalization and death [6]. A large population-based study in England reported 23,804 COVID-19–related deaths, of which one-third of the patients had diabetes (type 2 diabetes, n=7466, 31.4%; type 1 diabetes, n=365, 1.5%) [7].

Thus, the aim of this single-center cross-sectional study was to assess the clinical characteristics and outcomes of patients with diabetes admitted to a hospital in Dubai with moderate-to-severe COVID-19.

A total of 103 patients admitted to the hospital with a confirmed COVID-19 diagnosis had diabetes or prediabetes. During the same timeframe, overall, 410 patients with COVID-19 were admitted. Thus, 25.1% (103/410) of all patients admitted with COVID-19 had coexistent diabetes or prediabetes. Moreover, 66.9% (69/103) of the COVID-19 diabetes cohort were male patients. Patients within this study cohort represented 17 different ethnicities, of which 59.2% (61/103) were from Asian countries and 34.9% (36/103) were from Arab countries. The mean patient age was 54 (SD 12.5) years. In all, 73.8% (76/103) of the cohort patients had other comorbidities, including hypertension, ischemic heart disease, dyslipidemia, and chronic kidney disease (Table 1). Moreover, at admission, 9.7% (10/103) of the patients were receiving angiotensin converting enzyme inhibitor-I (ACE-I) treatment and 31.1% (32/103) patients were receiving angiotensin II receptor blocker (ARB) treatment, both of which were continued as per the individual patient’s clinical circumstances during admission, with no obvious adverse effects reported.

Out of the 103 cohort patients, 3 (2.9%) patients had type 1 diabetes and 90 (87.4%) patients had type 2 diabetes, 6 of whom were newly diagnosed with type 2 diabetes at admission. Moreover, 10 of 103 (9.7%) patients had prediabetes, 9 of whom were newly diagnosed at admission. Of those patients with a prior diagnosis of diabetes, 8 (7.8%) patients were on a basal insulin regimen and 6 (5.8%) patients were on basal-bolus insulin regimen (Table 2).

On admission, several laboratory investigations were performed (Table 3). Average random blood glucose was reported to be 10.2 mmol/L. Of the 103 patients; 56 (54.3%) had lymphopenia; 65 (63.1%) had received a confirmed diagnosis of pneumonia based on radiological findings; 38 (36.8%) had a normal chest radiograph, 14 of whom had confirmed pneumonia on chest HRCT; and 4 tested negative for COVID-19 based on laboratory RT-PCR but had radiologically confirmed COVID-19 pathognomonic changes.

Table 4 outlines emergent treatments used to manage patients with diabetes who were diagnosed with COVID-19 and their related outcomes. Of the 103 cohort patients, 50 (48.5%) patients received glucocorticoids at admission; 63 (61.2%) patients needed insulin at admission, requiring an average of 29.6 units per 24 h; and 5 (4.9%) patients had documented hypoglycemia (defined as blood glucose level <4 mmol/L) as detected by capillary blood glucose testing during admission (2 patients were on hydroxychloroquine, 4 on insulin, and 1 had end-stage renal disease).

Of all the patients (29/103, 28.2%) who were admitted to the intensive care unit (ICU), 21 were from Asian countries and 8, from Arab countries. Furthermore, 22 patients were male. In all, 5 (4.9%) patients (2 males, 3 females) died—all were in the ICU. Four of these patients were of Asian origin, 3 were known to have type 2 diabetes, and 2 had received a new diagnosis of prediabetes on admission. Moreover, 2 of those 5 patients were obese, 1 had end-stage renal disease with known breast cancer, all 5 had pneumonia, and 4 developed acute respiratory distress syndrome with septic shock. Their average length of stay in the hospital was 18 days.

Kaplan-Meier survival curve (Figure 1) showed that 4 of the 5 deaths that occurred in our study cohort were of patients who received both hydroxychloroquine and glucocorticoid treatment (as might be anticipated based on the severity of disease). The median time of death was 21 (range 3-39) days.

In this single-center study conducted in Dubai, 25.1% (103/410) of patients with COVID-19 already had diabetes or prediabetes at admission. Of the study cohort, 14.6% (15/103) received a new diagnosis of diabetes or prediabetes at admission. According to the International Diabetes Federation, the prevalence of diabetes in the United Arab Emirates is 15.4% [8], and our data suggests an overrepresentation of diabetes and prediabetes as risk factors for more severe COVID-19 illness requiring admission.

Most patients with diabetes or prediabetes admitted to hospital for COVID-19 (69/103, 66.9%) were male. This finding is in line with emerging studies that show men with COVID-19 are at a higher risk for developing severe outcomes, including death, than women [3,4,9].

Dubai is a highly multinational society comprising people of more than 200 nationalities. The patients in our study cohort represented 17 different nationalities, of which 59.2% (61/103) were from Asian countries and 34.9% (36/103) were from Arab countries. This finding is also consistent with other studies that suggest that people of Black, Asian, and Minority Ethnic (BAME) populations have increased risk and are predisposed to worse clinical courses and outcomes with COVID-19 than are their Caucasian counterparts [10]. In addition, many individuals of BAME origin are also more likely to have diabetes or prediabetes [11].

Moreover, consistent with recent studies, most patients in our cohort with diabetes or prediabetes and COVID-19 (76/103, 73.8%) also had other notable comorbidities. This finding is consistent with the fact that diabetes and cardiovascular disease are key components of the metabolic syndrome [12]. The metabolic syndrome is also associated with proinflammatory and prothrombotic states, which may have important implications for COVID-19 cases, wherein such complications are especially common and troublesome [13].

In this study, laboratory test results showed that fibrinogen, D-dimer, ferritin, and C-reactive protein (CRP) levels at admission were higher in patients subsequently requiring intensive care treatment. This finding is reflective of the inflammatory cytokine response observed in COVID-19. In addition, measuring admission glucose levels and in-hospital monitoring are important. Acute hyperglycemia shows upregulation of ACE2 gene, which facilitates entry of SARS-CoV-2 virus inside the cells. Prolonged hyperglycemia causes downregulation of ACE2 expression, making the cells vulnerable to the inflammatory effects of the SARS-CoV-2 virus [14]. Thus, performing these laboratory tests at hospital admission may be useful for risk stratification, that is, to determine the potential severity of the disease and guide the level of care that may be required for the patient.

The limitations of our study include the small sample size and retrospective nature of our analysis. Therefore, more studies with larger sample sizes are needed. Another limitation of our study is that there was no control group of patients without diabetes for comparison of results (other than knowing the absolute numbers of total patients admitted with COVID-19 during the same timeframe). In addition, because of the increasing number of new patients presenting with COVID-19 symptoms to our healthcare facility over a short timeframe, data on accurate characterization of overweight or obese status were not available for the majority of our patients; these conditions are thus not reported as comorbidities despite excess bodyweight being related to more severe COVID-19 illness and increased mortality [3,4]. Laboratory investigations such as fibrinogen, D-dimer, ferritin, and CRP levels can help in the initial assessment of severity of disease and subsequent need for ward-based or intensive care.