key: cord-0762399-01xyjaih
authors: Pal, Rimesh; Banerjee, Mainak; Yadav, Urmila; Bhattacharjee, Sukrita
title: Clinical profile and outcomes in COVID-19 patients with diabetic ketoacidosis: A systematic review of literature
date: 2020-08-18
journal: Diabetes Metab Syndr
DOI: 10.1016/j.dsx.2020.08.015
sha: 4f2aedc07b5fe9dff5d846d21219eee6e8ac207c
doc_id: 762399
cord_uid: 01xyjaih

BACKGROUND AND AIM: To perform a systematic literature review and analyze the demographic/biochemical parameters and clinical outcomes of COVID-19 patients with diabetic ketoacidosis (DKA) and combined DKA/HHS (hyperglycemic hyperosmolar syndrome). METHODS: PubMed, Scopus, Embase, and Google Scholar databases were systematically searched till August 3, 2020 to identify studies reporting COVID-19 patients with DKA and combined DKA/HHS. A total of 19 articles reporting 110 patients met the eligibility criteria. RESULTS: Of the 110 patients, 91 (83%) patients had isolated DKA while 19 (17%) had DKA/HHS. The majority of the patients were male (63%) and belonged to black ethnicity (36%). The median age at presentation ranged from 45.5 to 59.0 years. Most of the patients (77%) had pre-existing type 2 diabetes mellitus. Only 10% of the patients had newly diagnosed diabetes mellitus. The median blood glucose at presentation ranged from 486.0 to 568.5 mg/dl, being higher in patients with DKA/HHS compared to isolated DKA. The volume of fluid replaced in the first 24 h was higher in patients with DKA/HHS in contrast to patients with DKA alone. The in-hospital mortality rate was 45%, with higher mortality in the DKA/HHS group than in the isolated DKA group (67% vs. 29%). pH was lower in patients who had died compared to those who were discharged. CONCLUSION: DKA in COVID-19 patients portends a poor prognosis with a mortality rate approaching 50%. Differentiating isolated DKA from combined DKA/HHS is essential as the latter represents around one-fifth of the DKA cases and tends to have higher mortality than DKA alone.

People with diabetes mellitus (DM) represent a highly vulnerable population at a high-risk of poor prognosis with the novel coronavirus disease . The presence of DM increases the probability of severe disease, admission to the intensive care unit, and mortality due to COVID-19 (1) (2) (3) (4) . Impaired host defense, immune dysregulation with an increased predisposition to cytokine storm, and altered angiotensin-converting enzyme 2 (ACE2) expression have been implicated as the underlying pathophysiological mechanisms (3, 5) .

As with other severe infections (6) , diabetic ketoacidosis (DKA) has been reported in patients with COVID-19 (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) . With regard to coronaviruses, it has been shown that SARS-CoV (responsible for the SARS outbreak in 2003) binds to ACE2 in the pancreatic islets leading to islet damage and acute diabetes (25) . As SARS-CoV-2 (causative organism of COVID-19) also binds to ACE2, the virus might also result in acute diabetes (26) . This theoretical pathophysiology could lead to insulinopenia and increased risk of DKA, especially in patients with pre-existing DM (13). Besides, interleukin-6, an important cytokine of hyper-inflammatory state in COVID- 19 , has also been found to be elevated in DKA and serves as a driver of ketogenesis (27).

Thus, although there is insufficient data, DKA may be more prevalent in and SARS-CoV-2 may pose an increased risk over other severe infectious diseases (13). Apart from DKA, occasional cases of combined DKA and hyperglycemic hyperosmolar syndrome (HHS) have been reported in COVID-19 (22, 28) .

The clinical outcome of COVID-19 patients with DKA has been somewhat conflicting across studies; while one study found a mortality rate of 50% in COVID- 19 patients with DKA (9), another study found that patients with DKA were more likely to survive compared to non-DKA patients (17) . Moreover, patients with J o u r n a l P r e -p r o o f combined DKA and HHS tend to have higher mortality than either DKA or HHS alone (29), however, similar comparative data in COVID-19 is lacking.

The aim of the study was to provide a comprehensive systematic literature review of DKA and combined DKA/HHS in patients with confirmed COVID-19 in order to analyze the demographic and biochemical parameters and the clinical outcomes.

A systematic review of the literature was performed as per the PRISMA guidelines (30) across PubMed, Scopus, Embase and Google Scholar databases till August 3, 2020 using the following keywords: "COVID-19", "diabetic ketoacidosis, "ketosis", "ketonemia", "hyperglycemic emergencies", "hyperglycemic crises" with interposition of the Boolean operator "AND"/"OR". The search was conducted by two independent authors (RP and MB). Articles hence identified were further screened. Duplicate articles, articles in non-English language, reviews, and comments/communications and articles not pertaining to DKA in patients with COVID-19 were excluded. Finally, a total of 19 articles met the eligibility criteria and were included (Figure 1 ). The included articles and the number of patients in each article have been summarized in Table 1 .

The following data were extracted from the included studies: sex, age of the patient at presentation, ethnicity, type of diabetes mellitus (pre-existing T1DM vs. pre-existing T2DM vs. newly diagnosed DM), ongoing medications, body mass index (BMI), J o u r n a l P r e -p r o o f biochemical investigations (blood glucose, pH, bicarbonate, anion gap, glycated hemoglobin) and clinical outcomes.

Statistical analysis was performed using the Statistical Package for Social Sciences (SPSS) 23.0 software program (SPSS Inc., Chicago, IL, USA). Kolmogorov-Smirnov test was used to check the normality of individual patient data, wherever available.

Normally distributed data were expressed as mean ± standard deviation (SD), while non-parametric data were expressed in median (interquartile range, IQR). The comparison of biochemical parameters between groups (DKA vs. combined DKA and HHS; discharged vs. deceased) were made using Mann-Whitney U test. A p value < 0.05 was considered significant.

A total of 110 COVID-19 patients diagnosed with DKA were included in the final analysis. Amongst these 110 patients, 91 (83%) patients had DKA alone while 19 (17%) patients had combined DKA/HHS (22, 28) . The demographic data have been presented in Table 2 . Notably, the majority of the patients (63%) were male. The median age at presentation ranged from 45.5 years to 59.0 years. In the 28 patients in whom individual data was available, only one patient belonged to the pediatric age group (16) . The majority of the patients were black (African-American/Black African/African/Afro-Caribbean) (n=30, 36%), followed by Hispanic (n=19, 23%) and White (Caucasian) (n=10, 12%) ethnicity. The majority of the patients (77%) had pre-existing T2DM. The use of SGLT2 inhibitors was reported in 7 patients.

The biochemical parameters at admission have been summarized in Table 3 . The median blood glucose at presentation ranged from 486.0 mg/dl to 568.5 mg/dl. On separately analyzing patients in whom individual data were available (n=25), it was found that patients with combined DKA and HHS (n=6) had significantly higher blood glucose than those with DKA alone (n=22) (p=0.004). Three patients had blood glucose < 250 mg/dl at presentation (euglycemic DKA) (11,13,23); two were on SGLT2 inhibitor therapy (11,13) while one patient had gestational DM (23) . There was no difference in glycated hemoglobin (HbA 1c ) (p=0.225), pH (p=0.144), bicarbonate (p=0.242), and anion gap (p=0.478) between patients with DKA vs. those with combined DKA and HHS.

The majority of the patients were initially treated with a standard treatment protocol for DKA with intravenous fluids and insulin infusion. The median (IQR) volume of fluid replaced in the first 24 hours was 3.8 (3.0-5.0) liters and 5.0 (4.0-6.0) liters for patients with DKA and combined DKA/HHS, respectively (22) . In another series, the median (IQR) volume of fluid replaced was 3.0 (1.0-8.0) liters in the initial 24 hours (9). Only 8 patients reported in 2 studies were treated with subcutaneous insulin (9,13). Four patients had thromboembolic events (11, 15, 18, 20) . Data on the final outcome (in terms of discharged, deceased, or hospitalized) was available for 78 patients (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (18) (19) (20) (21) 23, 24, 28) . Among these 78 patients, 41 (52.5%) were discharged, 35 (45.0%) were deceased and 2 (2.5%) had remained hospitalized. Table   4 summarizes the clinical and biochemical predictors of final outcome in 27 patients (with individual patient data) who had either been discharged (n=17) or deceased (n=10) (7, 8, (10) (11) (12) (13) (14) (15) (16) (18) (19) (20) (21) 23, 24, 28) . All the deceased individuals were males. Patients J o u r n a l P r e -p r o o f with DKA/HHS had higher mortality compared to patients with isolated DKA (67% vs. 29%). In addition, deceased patients had lower pH at admission than those who were discharged (p=0.017).

To the best of our knowledge, this is the first comprehensive systematic review of DKA in patients with COVID-19. Amongst 110 cases hitherto described in the literature, 91 patients had DKA alone while 19 patients had combined DKA/HHS. The majority of the patients was male (63%) and had pre-existing T2DM (77%).

Apart from blood glucose, none of the biochemical parameters at presentation was significantly different between the DKA and DKA/HHS groups. The in-hospital mortality rate was 45%, with higher mortality in the DKA/HHS group than in the isolated DKA group (67% vs. 29%). pH was lower in patients who had died compared to those who were discharged.

Diabetic ketoacidosis is the most common hyperglycemic crisis, which also includes HHS and the combined syndrome of DKA and HHS, often referred to as hyperosmolar ketoacidosis (31). Diabetic ketoacidosis occurs in the setting of relative or absolute insulin deficiency that tips the insulin:glucagon ratio in favor of glucagon; this leads to reduced glucose utilization and unchecked lipolysis, causing excessive formation of ketone bodies and finally metabolic acidosis (32,33). Although DKA is more likely to occur in T1DM, it is estimated that the majority of the DKA cases worldwide occur in patients with T2DM due to its higher prevalence (33). Diabetic There are theoretical concerns that the SARS-CoV-2, just like its cousin the SARS-CoV, could bind to ACE2 expressed on the pancreatic islets, leading to islet destruction and acute diabetes (25, 26) . In a patient with underlying DM (especially T2DM), the destruction of the residual β-cells by the virus would result in a state of complete insulinopenia, thereby precipitating DKA (13). Moreover, the proinflammatory milieu seen even in non-severe patients with COVID-19 would theoretically promote ketogenesis, hence, predisposing an individual to ketosis (27).

In a series of 658 hospitalized patients (129 had DM) with confirmed COVID-19 from China, 42 (6.4%) presented with either urine or serum ketones. The presence of ketosis was associated with higher rates of acute respiratory distress syndrome, acute liver injury, the requirement for mechanical ventilation, a longer length of hospital stay, and mortality. Of the 42 patients with ketosis, 15 (35.7%) had diabetes mellitus; of these 15 patients, 3 had concomitant acidosis, amounting to DKA. Hence, the overall prevalence of DKA amongst patients with DM in the series was 2.3% (21) . On the contrary, in a retrospective single-center study from the United Kingdom, the prevalence of DKA in 87 COVID-19 patients with DM was 9.2% (17) . In another series from the United Kingdom, the prevalence of DKA in confirmed COVID-19 patients was 5.9% (10). The marked dissimilarity in the prevalence of DKA may be explained based on differences in the population studied and the underlying COVID-19 disease severity.

In our systematic review, we found that the majority (63%) of the reported cases of DKA (or DKA/HHS) were male. Gender difference as a risk factor for DKA is controversial; some studies consider female sex as a risk factor (37) , while others shown have no significant differences in males and females (38) . The preponderance J o u r n a l P r e -p r o o f of male cases among DKA patients could be due to the fact that COVID-19 tends to be more severe in males, resulting in more hospitalizations and deaths compared with females (39, 40) . Similarly, the relative absence of pediatric COVID-19 patients with DKA also underscores the fact that children tend to have lower rates of severe COVID-19 (41) . Regarding ethnicity, we found that 36% of the patients belonged to black ethnicity, while 23% were Hispanic, 12% were Caucasian and only 7% were Asian. In general, BAME (Black, Asian and Minority Ethnic) individuals are at an increased risk of acquiring SARS-CoV-2 infection and of worse clinical outcomes from COVID-19 compared to White individuals (42) . In addition, ethnic minorities are also at a higher risk of DKA (43) . Taken together, the high prevalence of DKA in people with black ethnicity is ably explained. On the contrary, the underrepresentation of the Asian population could be fallacious as all the studies reporting patient ethnicity were either from the United Kingdom or from the United States of America (9,10,22,28).

Regarding the type of DM, we found that the majority of COVID-19 patients with DKA (77%) had underlying T2DM, likely because of the higher prevalence of T2DM worldwide (33). In 10 patients, DM was diagnosed at admission, categorizing them as newly diagnosed DM. Of these 10 patients, 7 had HbA1c > 9.5% (ranging from 9.6% to 14.2%) (10,12,13,16,24,28), implying that these patients had underlying 

Diabetes in COVID-19: Prevalence, pathophysiology, prognosis and practical considerations

Diabetes mellitus is associated with increased mortality and severity of disease in COVID-19 pneumonia -A systematic review, meta-analysis, and meta-regression

COVID-19, Diabetes Mellitus and ACE2: The conundrum

COVID-19 and diabetes mellitus: An unholy interaction of two pandemics

COVID-19 in people with diabetes: understanding the reasons for worse outcomes

DKA cases over the last three years: has anything changed?

Newly diagnosed diabetes and diabetic ketoacidosis precipitated by COVID-19 infection

COVID-19 symptoms masking inaugural ketoacidosis of type 1 diabetes

COVID-19 Presenting as Acute Limb Ischaemia

COVID-19 in a Child With Diabetic Ketoacidosis: An Instigator, a Deviator or a Spectator

Clinical characteristics and outcomes of COVID-19 hospitalized patients with diabetes in the United Kingdom: A retrospective single centre study

Cerebral Venous Thrombosis Associated with COVID-19

Acute Hyperglycemic Crises with Coronavirus Disease-19: Case Reports

A 46-Year-Old Woman Who Presented with Diabetic Ketoacidosis and COVID-19 Pneumonia with Multiple Pulmonary Thromboemboli: A Case Report

COVID -19 infection may cause ketosis and ketoacidosis

Protracted ketonaemia in hyperglycaemic emergencies in COVID-19: a retrospective case series

Euglycaemic ketoacidosis during gestational diabetes with concomitant COVID-19 infection

Diabetic ketoacidosis precipitated by COVID-19: A report of two cases and review of literature

Binding of SARS coronavirus to its receptor damages islets and causes acute diabetes

COVID-19 and the endocrine system: exploring the unexplored

National survey of the management of Diabetic Ketoacidosis (DKA) in the UK in 2014

Diabetic emergencies -ketoacidosis, hyperglycaemic hyperosmolar state and hypoglycaemia

Diabetic Ketoacidosis: Risk Factors and Management Strategies

Risk factors and prevention strategies for diabetic ketoacidosis in people with established type 1 diabetes

Event Rates and Risk Factors for the Development of Diabetic Ketoacidosis in Adult Patients With Type 1 Diabetes: Analysis From the DPV Registry Based on 46,966 Patients

Biological sex impacts COVID-19 outcomes

Impact of sex and gender on COVID-19 outcomes in Europe

The impact of ethnicity on clinical outcomes in COVID-19: A systematic review

Rates of Diabetic Ketoacidosis: International Comparison With

Pediatric Patients With Type 1 Diabetes From England

Serological Prediction of infections in Diabetic Patients with Diabetes Ketoacidosis in

Hyperglycemic crises in urban blacks

Characteristics and outcomes of the hyperglycemic hyperosmolar nonketotic syndrome in a cohort of 51 consecutive cases at a single center

Joint British Diabetes Societies guideline for the management of diabetic ketoacidosis: Diabetic ketoacidosis guidelines

Hyperglycemic Crises in Adult Patients With Diabetes

Management of critically ill patients with COVID-19 in ICU: statement from front-line intensive care experts in Wuhan, China. Ann Intensive Care

Guidance on the management of Diabetic Ketoacidosis in the exceptional circumstances of the COVID-19 pandemic

Management of Hyperglycemic Crises in Patients With Diabetes

Subcutaneous rapid-acting insulin analogues for diabetic ketoacidosis

Remdesivir in adults with severe COVID-19: a randomised, double-blind, placebo-controlled, multicentre trial. The Lancet

Assessment of Hypokalemia and Clinical Characteristics in Patients With Coronavirus Disease

Disturbance of inorganic phosphate metabolism in diabetes mellitus: clinical manifestations of phosphorus-depletion syndrome during recovery from diabetic ketoacidosis

Trends in Diabetic Ketoacidosis Hospitalizations and In-Hospital Mortality -United States

Incidence and long-term outcomes of critically ill adult patients with moderate-to-severe diabetic ketoacidosis: Retrospective matched cohort study

COVID-19 coagulopathy: an evolving story

COVID-19-associated Hemophagocytic Lymphohistiocytosis and Coagulopathy: Targeting the Duumvirate. Indian Pediatr

^ Only data of patients with isolated DKA has been shown @ Individual data on patient sex were available in 102 patients from 18 studies

Patients with black ethnicity include African-American, Black African, African and Afro-Caribbean patients ¶ Individual data on type of diabetes were available in 97 patients from 15 studies

$ Use of SGLT2 inhibitors was reported in 7 patients from 5 studies

¥ BMI calculated from individual patient data available from 6 studies COVID-19: Novel coronavirus disease; DKA: Diabetic ketoacidosis; HHS: Hyperglycemic hyperosmolar syndrome

HbA 1c (%) 11.7 (9

Anion gap (mEq/l)