key: cord-0967266-u2v1w99n
authors: Korytkowski, Mary; Antinori-Lent, Kellie; Drincic, Andjela; Hirsch, Irl B; McDonnell, Marie E; Rushakoff, Robert; Muniyappa, Ranganath
title: A Pragmatic Approach to Inpatient Diabetes Management during the COVID-19 Pandemic
date: 2020-06-04
journal: J Clin Endocrinol Metab
DOI: 10.1210/clinem/dgaa342
sha: f4d3a7a68c569e27e15db858e76438d3af81eb1e
doc_id: 967266
cord_uid: u2v1w99n

The pandemic of COVID-19 has presented new challenges to hospital personnel providing care for infected patients with diabetes who represent more than 20% of critically ill patients in intensive care units. Appropriate glycemic management contributes to a reduction in adverse clinical outcomes in acute illness but also requires intensive patient interactions for bedside glucose monitoring, intravenous and subcutaneous insulin administration, as well as rapid intervention for hypoglycemia events. These tasks are required at a time when minimizing patient interactions is recommended as a way of avoiding prolonged exposure to COVID-19 by health care personnel who often practice in settings with limited supplies of personal protective equipment. The purpose of this manuscript is to provide guidance for clinicians for reconciling recommended standards of care for infected hospitalized patients with diabetes while also addressing the daily realities of an overwhelmed health care system in many areas of the country. The use of modified protocols for insulin administration, bedside glucose monitoring, and medications such as glucocorticoids and hydroxychloroquine that may affect glycemic control are discussed. Continuous glucose monitoring systems have been proposed as an option for reducing time spent with patients, but there are important issues that need to be addressed if these are used in hospitalized patients. On site and remote glucose management teams have potential to provide guidance in areas where there are shortages of personnel who have expertise in inpatient glycemic management.

212 mg/dL (11.8 mmol/L). An HbA1c done the month prior to admission was 7.6% (60 mmol/mol). His electrolytes and liver function tests were normal. Real-time reverse transcription polymerase chain reaction (RT-PCR) assay on a nasopharyngeal swab was positive for SARS-CoV-2. Chest radiographs revealed bilateral infiltrates.

The patient was admitted and started on azithromycin and hydroxychloroquine. His home diabetes medications were discontinued and he was started on glargine insulin with a correction insulin scale prior to meals. Point of care blood glucose (POC BG) levels ranged between 140 and 210 mg/dl. Two days after the admission, he experienced worsening hypoxia. He was placed on 50% FIO2 and transferred to the ICU where he was placed on mechanical ventilation due to worsening respiratory failure and hypotension.

A c c e p t e d M a n u s c r i p t 5 The pandemic of COVID-19 has presented new challenges to hospital personnel providing care for these patients. This is particularly true for patients with diabetes, who represent 25 to 34% of the patient population receiving care in intensive care unit (ICU) and non-ICU settings and for whom appropriate glycemic management may contribute to a reduction in adverse clinical outcomes (1) (2) (3) (4) . Attention to glycemic management reduces morbidity and mortality in hospitalized patients with diabetes or newly recognized hyperglycemia with acute illness, including those with the SARS and COVID-19 virus (3, (5) (6) (7) .

If left untreated, hyperglycemia increases risk for infections by altering leukocyte function and increasing the virulence of some pathogens; enhances risk for cardiac arrhythmias; prolongs hospital length of stay; and increase mortality (5,8). Implementation of protocols to control BG levels while also avoiding hypoglycemia have the ability to reduce these adverse outcomes (8-10).

Current recommendations for inpatient glycemic management include frequent monitoring of bedside BG together with structured insulin regimens. Insulin regimens composed of long acting and short or rapid acting insulin preparations are recommended for achieving glycemic targets of 100-180 mg/dl in non-critically ill patients (5,6). Intravenous (IV) insulin infusions are recommended for achieving glycemic targets of 140-180 mg/dl in critically ill patients (5,6,11).

The ability to safely achieving these goals in the current environment has been viewed as problematic as glycemic management requires frequent patient interactions at a time when limiting these encounters is recommended, particularly in areas where there are shortages of personal protective equipment (PPE). These conflicting challenges have resulted in frequent queries from hospital personnel as to how to meet recommended standards of care for hospitalized patients with diabetes infected with SARS-CoV-2 while also addressing the daily realities of an overwhelmed health care system in many areas of the country.

A c c e p t e d M a n u s c r i p t 6 Some suggested methods for limiting exposure time for health personnel when caring for patients with COVID-19 includes minimizing the use of IV insulin infusions in critically ill patients, using remote continuous glucose monitoring devices (CGM) devices to minimize time spent in direct patient contract, and reconsidering use of non-insulin therapies. In addition, the role of diabetes self-management by patients with diabetes in the hospital has gained renewed interest (11) (12) (13) (14) . Many hospitals are implementing strategies for limiting patient interactions, some of which may be achieved at risk of more hyperglycemia (15) .

The purpose of this communication is to provide guidance for clinicians managing hospitalized patients with COVID-19 and diabetes or newly recognized hyperglycemia while also addressing the needs for protecting personnel who interact with these patients (16) . It should be noted that much of this discussion will not be based on randomized controlled clinical trials for patients with COVID-19, but is instead extrapolated and modified from prior evidence-based guidelines for inpatient glycemic management as well as from the clinical experiences of several of the authors providing care to these patients. It is generally recommended that hospitals not make major changes to their current approach to managing hospitalized COVID-19 patients with hyperglycemia due to concerns that this alone can increase risk for unintended consequences requiring more time at the bedside. However, there are important and emerging issues that directly affect established glycemic management that warrant discussion and consideration during this pandemic (16) .

Two groups of patients may fall into this category, those with well-controlled non-insulin treated type 2 diabetes and those with newly recognized hyperglycemia, defined as verified BG > 180 mg/dl (10 mmol/L). These patients require POC BG monitoring with initial use of correction A c c e p t e d M a n u s c r i p t 7 insulin to achieve and maintain BG between 100 and 180 mg/dl (5.5-10 mmol/L). Measurement of an HbA1c on admission helps identify patients with previously undiagnosed diabetes (17) . A c c e p t e d M a n u s c r i p t 8 mg/dl (10 mmol/L) will require regular glucose monitoring together with scheduled insulin therapy to achieve desired glycemic goals. Scheduled insulin therapy is defined as the use of a basal insulin administered preferably as a long acting insulin preparation (e.g. glargine U100), prandial insulin for patients who are eating or receiving supplemental enteral or parenteral nutrition, and correction insulin for BG above target range ( care settings as a way of minimizing direct patient interactions. Administering meal related rapid acting insulin following a meal has been advocated for some hospitalized patients who have reduced food intake (33). However, this practice necessitates additional nurse time with a patient, and has not been consistently observed to reduce hypoglycemia events (34,35).

Previously published guidelines recommend discontinuation of non-insulin medications and initiation of insulin therapy for patients with diabetes or newly recognized hyperglycemia at time of hospital admission (5,6,11). There are several more recent studies investigating the safety and efficacy of these earlier recommendations as well as use of non-insulin agents in the hospital (36,37). This emerging data for use of non-insulin therapies in the hospital setting will be addressed here in reference to COVID-19 patients (Table 1 ).

There are several small studies demonstrating safety and efficacy of the dipeptidyl There are major issues associated with use of other non-insulin therapies in the hospital ( A c c e p t e d M a n u s c r i p t 12 Several GLP1RA are now available as once weekly formulations, which means that many hospitalized patients will have this on board at time of admission. Similar to DPP4i, continued use of these agents is generally not recommended for acutely ill patients with COVID-19 due to the potential for abrupt deterioration in clinical status.

There has been an expanded use of SGLT2 inhibitors in the outpatient setting due to their beneficial effects on cardiovascular and renal outcomes in people with and without diabetes (50).

This means that many patients will be on these agents at the time of hospitalization. There is no data guiding their use in the inpatient setting, but the association of medications in this class with risk for euglycemic diabetic ketoacidosis (euDKA) and volume depletion contraindicates their use in any inpatient at this time (51,52). Consideration for stopping these agents in outpatients who become ill with COVID-19 is recommended given that glucosuria continues for several days following discontinuation, increasing risk for euDKA (53).

In summary, insulin therapy remains the standard of care for management of hyperglycemia in patients hospitalized with COVID-19. Selected use of the DPP4i, sitagliptin and linagliptin, can be considered for patients with type 2 diabetes or milder degrees of hyperglycemia once they are eating regular meals and discharge to home is anticipated. Patients receiving sitagliptin require monitoring of renal function with adjusted doses for renal insufficiency (40).

At this time, despite the extensive publications demonstrating CVD and renal benefits with use of SGLT2i, these agents require discontinuation on admission (or even earlier) for any patient with COVID-19 due to concerns for euDKA, infection and hypovolemia.

Bedside BG monitoring using POC glucose meters remains the standard of care in hospitalized patients, despite identified issues relating to accuracy and precision of several of these A c c e p t e d M a n u s c r i p t 13 devices (11, 54 There are no CGM devices that are approved for inpatient use. Of the several small trials allows glycemic data to be remotely transmitted to a receiver located outside a patient room as well as to computers at a nursing station.

It is important to note that manufacturers of the two CGM devices with temporary FDA allowances recommend against using sensor data for making treatment decisions related to insulin therapy (61 Non-critically ill patients using CGM as outpatients may be permitted to continue these devices as inpatients if the same considerations relating to radiologic procedures and use of certain medications that can potentially interfere with results are followed. Some hospitals have protocols in place for individuals using these monitoring devices in the non-critical care hospital setting (61,62).

Diabetes self-management, defined as allowing selected patients to monitor BG and administer their own insulin, may be appropriate for patients who are knowledgeable, competent and clinically stable (11, 14) . Patients using continuous subcutaneous insulin infusion (CSII) therapy prior to admission are often inappropriately instructed to discontinue this at time of hospital admission (65-68). Patients treated with CSII assessed as competent to continue this in A c c e p t e d M a n u s c r i p t 16 the hospital setting need to provide their own pump supplies including infusion sets, cartridges, reservoir syringes and batteries. Many CSII patients own systems that integrate with CGM. There are currently two products that allow insulin adjustment based on algorithms that "learn" glucose patterns. Questions have been raised as to whether these hybrid closed loop insulin systems (Medtronic 670G, Tandem's Control IQ program), are appropriate for use in the inpatient setting.

In light of the reports of high glycemic variability during COVID-19 and treatment, and especially rapid decline in insulin needs that can occur with improvement, it is recommended that these programs be disabled. For those who continue using their hybrid closed loop system, hospital protocols should be followed to prevent untoward misadventures as can occur when exposed to an electromagnetic field, radiation or some medications (discussed above).

Patients assessed as being able to self-monitor POC BG and administer their own insulin using CSII or insulin injections need to do so according to hospital procedures and guidelines that require reporting of glycemic data and insulin dosing to nursing personnel who can record this in the EMR. These policies typically indicate that insulin dose calculations and administration be overseen by nursing personnel.

Patients may self-monitor BG levels using their own home BG meter, or a meter provided by the hospital. In the latter case, the meter is restricted to use only by one individual patient. In addition to recent guidance regarding inpatient CGM use, the FDA has recognized that home BG meters may be used by some hospitalized patients with diabetes and COVID-19 (69).

It is essential that non-critically ill patients performing self-management be reassessed several times a day given observed rapid deteriorations in clinical status with COVID-19 (2).

Patients using CSII therapy who experience a deterioration in their clinical condition requiring transition to SC or IV insulin therapy will need to have these devices removed and given to a A c c e p t e d M a n u s c r i p t 17 family member or placed in a secured area until this can be returned to the patient. The infusion catheter for the device also needs to be removed from the skin to avoid a potential source of infection.

There are multiple drug therapies being studied using unique investigative strategies for determining optimal therapies for treating COVID-19 patients (70) . Of these, systemic glucocorticoid therapy and hydroxychloroquine are agents with divergent but significant impact on glycemic control in patients with and without diabetes. 

Hydroxychloroquine is an agent under investigation for COVID-19 that has relevance to glycemic management (20). In one small trial, therapy with hydroxychloroquine resulted in improved beta cell function and insulin sensitivity in obese subjects with insulin resistance but without diabetes (79) . In a systematic review of 18 studies with 55,776 participants, substantial reductions in were observed for HbA1c, fasting and postprandial plasma glucose levels. Insulin dose reduction by ~30% may be required for some patients receiving hydroxychloroquine to avoid hypoglycemia (80, 81) .

There are other medications used in hospitals (as well as outpatient settings) for patients with diabetes who are infected with COVID-19 that can affect blood glucose levels. An example of this would be the use of antitussive syrups containing glucose that can contribute to hyperglycemia.

Given the number of patients hospitalized with COVID-19 who also have diabetes or hyperglycemia, any inpatient diabetes service can be quickly overwhelmed with requests for consultation for glycemic management. In general, non-critically ill patients who have persistent BG < 180 mg/dl do not require subspecialty consultation. Both IV and SC insulin therapy can be A c c e p t e d M a n u s c r i p t 19 initiated for patients with persistent BG > 180 mg/dl according to hospital or published guidelines (5,6) ( Table 2) .

Glucose management teams may still need to be involved in the care of many patients with COVID-19, including those using CSII or who present difficult glycemic management issues despite implementation of guidelines described above. This can often be performed in the context of an e-consult or virtual telemedicine consult, with the goal of minimizing the number of personnel coming into direct contact with any one patient (82, 83) .

collate glucose, insulin, nutrition and medication information in the EMR for remote review by a diabetes management team who makes remote recommendations to the primary team for adjustments in therapy (82, 83) . Implementation of a vGMS with daily suggestions for patients with elevated POC BG values resulted in a nearly 40% decrease in percentage of patients with hyperglycemia (83) . For COVID-19 patients with glycemic management issues, a vGMS could be adapted to an active patient Dashboard to allow remote monitoring with suggestion for interventions as needed (84) .

Many but not all patients with diabetes hospitalized with COVID-19 infection will be knowledgeable regarding self-management at home. Even those who were previously comfortable with home management may be discharged with a different regimen than they were using prior to admission (85) . Diabetes education and training is a key part of comprehensive diabetes care and should remain a part of discharge planning in the COVID-19 pandemic (85) . Continuing to take advantage of technology, delivery of patient education can continue using telehealth with a HIPPA A c c e p t e d M a n u s c r i p t 20 compliant platform using tablets, computers, or smart phones (86, 87) . Bluetooth enabled pen devices can allow remote monitoring of compliance with timing and dosing of insulin regimens and identify patients who are at risk for uncontrolled diabetes (88). All self-management education should begin well before the day of discharge. Patients new to insulin should have the opportunity to practice self-administration using devices (vials and syringes, pen devices) they will use at home. Patients need to know how and when to take their diabetes medications, monitor POC BG levels, adjust therapy for low or high BG values, and who to contact in the event of glycemic emergencies. All patients discharged home with insulin or an insulin secretagogue need to know the symptoms and treatment of hypoglycemia events. For patients receiving basal bolus insulin therapy, a prescription for nasal or injectable glucagon provides reassurance that they will have appropriate tools in the event of a severe hypoglycemic reaction.

Following transfer to the ICU and intubation, the patient was started on IV insulin using a protocol targeting BG values of 140-180 mg/dl (7.8-10 mmol/L 

Hospitalization Rates and Characteristics of Patients Hospitalized with Laboratory-Confirmed Coronavirus Disease 2019 -COVID-NET, 14 States

Clinical Characteristics of Covid-19 in

Outcomes in Patients With Hyperglycemia Affected by Covid-19: Can We Do More on Glycemic Control? Diabetes Care

Presenting Characteristics, Comorbidities, and Outcomes Among

Intensive versus conventional glucose control in critically ill patients

Diabetes Care in the Hospital: Standards of Medical Care in Diabetes -2020

Patient Self-Management of Diabetes Care in the Inpatient Setting: Pro

Patient Self-Management of Diabetes Care in the Inpatient Setting: Con

Self-management of diabetes in hospital: a guideline from the Joint British Diabetes Societies (JBDS) for Inpatient Care group

COVID-19 in patients with diabetes: risk factors that increase morbidity

Our Response to COVID-19 as Endocrinologists and Diabetologists

Prevalence of undiagnosed and suboptimally controlled diabetes by point-of-care HbA1C in unselected emergency department patients

Results of a Study Comparing Glycated Albumin to Other Glycemic Indices

Consensus Statement on Use of Continuous Subcutaneous Insulin Infusion Therapy in the Hospital

FAQs on Home-use Blood Glucose Meters Utilized Within Hospitals During the COVID-19 Pandemic

The Randomized Embedded Multifactorial Adaptive Platform for Community-acquired Pneumonia (REMAP-CAP) Study: Rationale and Design

Prevalence, predictors, and outcomes of steroid-induced hyperglycemia in hospitalized patients with hematologic malignancies

Insulin requirements in non-critically ill hospitalized patients with diabetes and steroid-induced hyperglycemia

COVID-19: consider cytokine storm syndromes and immunosuppression

Hospital insulin protocol aims for glucose control in glucocorticoid-induced hyperglycemia

Use of NPH insulin for glucocorticoid-induced hyperglycemia

Pilot study of using neutral protamine Hagedorn insulin to counteract the effect of methylprednisolone in hospitalized patients with diabetes

Safety and efficacy of continuous insulin infusion in noncritical care settings

Implementing an intravenous insulin infusion protocol in the intensive care unit

Antidiabetogenic effects of hydroxychloroquine on insulin sensitivity and beta cell function: a randomised trial

Potential Effect of Hydroxychloroquine in Diabetes Mellitus: A Systematic Review on Preclinical

Case report of chloroquine therapy and hypoglycaemia in type 1 diabetes: What should we have in mind during the COVID-19 pandemic?

Remote Monitoring and Consultation of Inpatient Populations with Diabetes

Association between a virtual glucose management service and glycemic control in hospitalized adult patients: An observational study

Common Models Used for Inpatient Diabetes Management

Practical Recommendations for Transitioning Patients with Type 2 Diabetes from Hospital to Home

Telemedicine for Reach, Education, Access, and Treatment (TREAT): linking telemedicine with diabetes self-management education to improve care in rural communities

Remote Patient Monitoring and Clinical Outcomes for Postdischarge Patients with Type 2 Diabetes

A c c e p t e d M a n u s c r i p t 32 A c c e p t e d M a n u s c r i p t 35

Munshi MN, Slyne C, Greenberg JM, Greaves T, Lee A, Carl S, Atakov-Castillo A, A c c e p t e d M a n u s c r i p t 36 Majority of inpatient studies with these agents used these in combination with correction or basal insulin.Generally not recommended in acute phase of COVID-19 due to concerns for abrupt deteriorations in clinical status.Saxagliptin and alogliptin should not be used as they are associated with higher risk for HF. Administer rapid acting or regular insulin prior to administration of enteral nutrition (similar to patients eating meals). Some patients may also require basal insulin Administered prior to bolus *Insulin doses require daily (or more frequent) adjustments to achieve glycemic goals without hypoglycemia **Patients with diabetes and COVID-19 will likely require higher insulin doses based on the severity of the underlying insulin resistance. Many may require well over 1 units/kg/day of insulin during acute phase of illness. There are some patients, such as those with chronic kidney disease or who experience acute kidney injury who may require lower insulin doses to avoid hypoglycemia. ¶ The dose of prandial insulin will vary according to the type of formulation used. For patients with diabetes, a starting dose of 1 unit for every 15 to 20 grams of carbohydrate administered over 24 hours A c c e p t e d M a n u s c r i p t 38 could be calculated and administered in divided doses as a rapid acting insulin analog or regular insulin. For patients with new hyperglycemia, a correction insulin scale could be used initially to determine the need for ongoing scheduled prandial insulin coverage. ± Insulin requirements may be lower in patients receiving low carbohydrate enteral nutrition formulations (Reference #20). In the event of abrupt discontinuation of enteral nutrition in insulin treated patients, a 10% dextrose infusion administered at the same rate is recommended for the duration of the longest acting insulin administered prior to discontinuation (Reference 6).

Give 50-60% of TDD as basal insulin Give 40-50% of TDD as bolus (premeal or nutritional) in Use Correction Insulin (SSI) for BG above goal rang