Background

Since December 2019, an outbreak of coronavirus disease, officially named by the World Health Organization as COVID-19, appeared in Wuhan, Hubei Province, China. Diabetic patients infected with SARS-CoV-2 appear more prone to require hospitalization and admission to intensive care unit.

Transient non-severe increased insulin requirement in patients hospitalized for various medical conditions such as sepsis or myocardial infarction is a well-known phenomenon with insulin doses usually not exceeding 5–10 Units/hours [1]. However, extremely high-dose insulin requirement remains a very rare reported entity. Here, we report the case of an extreme and transitory high-dose insulin requirement episode in a diabetic patient presenting an acute respiratory distress syndrome (ARDS) caused by SARS-CoV-2.

Case presentation

A 57-year-old man previously known for type 2 diabetes (T2D) for 3 years, asthma and obstructive sleep apnea was admitted for an aggravation of his dyspnea. Bodyweight was 99 kg, height was 172 cm and body mass index was 33.5 kg/m2.

His T2D was treated only with metformin 500 mg twice daily and his latest Hb1Ac was 6.1%. The patient was not on other regular treatment. In the past 3 weeks, he presented a dry cough that was initially managed with amoxicillin/clavulanic acid and clarithromycin for 5 days. At admission, he presented a normal leucocyte count at 4.3 G/l with lymphopenia at 0.39 G/l count and an elevated CRP at 58 mg/l [2, 3]. Chest CT SCAN showed a bilateral multilobar ground-glass opacification. A nasopharyngeal swab test of SARS-CoV-2 by RT-qPCR was performed the same day in another medical facility and the result was pending.

Twenty-four hours after his admission he presented a worsening of dyspnea, with tachypnea and severe hypoxemia requiring a transfer to the intensive care unit rapidly followed by oro-tracheal intubation for mechanical ventilation support. A vasopressor infusion (norepinephrine) was started. A bronchoalveolar lavage was performed and test of SARS-CoV-2 by RT-qPCR assay was positive. At day 3, he presented a rapidly progressive insulin requirement at a rate of up to 50 units/hour iv of insulin aspart (Table 1). Despite the high insulin doses, he maintained an elevated plasma glucose level at 270 mg/dL on average.

Table 1 clinical and biological parameters during intensive care unit stay
Full size table

His extremely high-dose insulin requirement “resolved” at day 9, and the insulin infusion rate was rapidly reduced, avoiding hypoglycemia (Table 2). He was extubated at day 13 and discharged to an intermediate care unit as a step down approach to pursue a close monitoring and regular respiratory care including continuous positive airway pressure (CPAP), and 2 days later he was transferred to regular hospital room due a positive clinical evolution. His insulin treatment was switched to subcutaneous neutral protamine Hagedorn (NPH) insulin at a total dose of 60 UI per day.

Table 2 Laboratory values at intravenous insulin initiation and withdrawal
Full size table

Discussion and conclusions

This case could reflect a potential severe impact of COVID-19 infection on metabolic homeostasis in a sub-group of patients, in particular diabetic individuals that appear more prone to complications of COVID-19 infection. Yet, the mechanisms behind this remain to be elucidated. No previous report has been made in COVID-19 infection of extremely high-dose insulin requirement. Available data remain limited to assess if COVID-19 infections are more prone to induce metabolic disturbances than SARS-CoV, but previous observations with SARS in 2003 showed that both hyperglycemia and history of diabetes were independent predictors of severe respiratory symptoms and death in SARS-CoV patients, as observed with COVID-19 infection [4,5,6,7].

The metabolic response to sepsis encompasses several phenomenons that can contribute to hyperglycemia such as the activation of the hypothalamic-pituitary-adrenal axis, leading to the secretion of cortisol, growth hormone and glucagon resulting in an increased level of glucose. In addition, several mediators of the systemic inflammatory response such as TNF-α or IL-1 may cause hepatic and muscle insulin resistance by disturbing insulin receptor signaling [8]. Interestingly COVID-19 infected patients often present normal or low leucocytes counts and lymphopenia as in our patient [9]. Furthermore, endogenous and exogenous catecholamine used as vasopressors have been reported to increase glucose levels by stimulating glycogenolysis, gluconeogenesis, inhibiting glycogen synthesis and stimulating the release of growth hormone and glucagon [10]. Moreover, a study showed that SARS-CoV may cause diabetes by ACE2-dependent damage of pancreatic islet cells [11]. and similar observation have been made in patients infected with SARS-CoV2, suggesting that this mechanism could be one the factors contributing to metabolism homeostasis disturbance leading to new-onset diabetes and acute hyperglycemia in COVID-19 infection [12].

In the present case, we could hypothesize that COVID-19 infection has led to a high insulin resistance state through several mechanisms including a large release of cytokines named cytokine storm as described in several reports. In addition, catecholamine administration may have exacerbated this state [13]. Furthermore, T2D is often associated with insulin-resistance due to a low-grade inflammatory state and abnormal plasma oxidative stress that can be exacerbated during an infection episode, leading to intense insulin resistance as presented by our patient [8].

Observational studies are required to better elucidate the impact of COVID-19 infection on glucose metabolism, in particular in diabetic patients, by several approaches such as comparing insulin requirement between patients with COVID-19 infection and patients with other causes of septic shock and ARDS to eventually confirm a potential disproportionate insulin requirement during COVID-19 infection.