Diabetes Ther https://doi.org/10.1007/s13300-021-01114-x REVIEW Telemonitoring, Telemedicine and Time in Range During the Pandemic: Paradigm Change for Diabetes Risk Management in the Post-COVID Future Thomas Danne . Catarina Limbert . Manel Puig Domingo . Stefano Del Prato . Eric Renard . Pratik Choudhary . Alexander Seibold Received: May 27, 2021 / Accepted: July 1, 2021 Ó The Author(s) 2021 ABSTRACT including adults and children with T1D, allow- ing an exploration of the impact of COVID-19 People with diabetes are at greater risk for neg- inside and outside the hospital setting on gly- ative outcomes from COVID-19. Though this caemic control. To date, 27 studies including risk is multifactorial, poor glycaemic control 69,294 individuals with T1D have reported the before and during admission to hospital for effect of glycaemic control during the COVID- COVID-19 is likely to contribute to the 19 pandemic. Despite restricted access to dia- increased risk. The COVID-19 pandemic and betes clinics, glycaemic control has not deteri- restrictions on mobility and interaction can also orated for 25/27 cohorts and improved in 23/27 be expected to impact on daily glucose man- study groups. Significantly, time in range (TIR) agement of people with diabetes. Telemonitor- 70–180 mg/dL (3.9–10 mmol/L) increased ing of glucose metrics has been widely used across 19/27 cohorts with a median 3.3% during the pandemic in people with diabetes, (- 6.0% to 11.2%) change. Thirty per cent of E. Renard Department of Endocrinology, Diabetes, Nutrition, T. Danne (&) Montpellier University Hospital, Montpellier, Diabetes Center for Children and Adolescents, France Kinder- und Jugendkrankenhaus AUF DER BULT, Janusz-Korczak-Allee 12, 30173 Hannover, Germany E. Renard e-mail: danne@hka.de Department of Physiology, Institute of Functional Genomics, CNRS, INSERM, University of C. Limbert Montpellier, Montpellier, France Unit for Paediatric Endocrinology and Diabetes, CHULC, Hospital Dona Estefania, Lisbon, Portugal P. Choudhary Department of Diabetes and Nutritional Sciences, C. Limbert King’s College London, London, UK NOVA Medical School, Lisbon, Portugal P. Choudhary M. Puig Domingo Diabetes Research Centre, University of Leicester, Endocrinology and Nutrition Service, Department Leicester, UK of Medicine, Germans Trias i Pujol Research Institute and Hospital, Universitat Autònoma de A. Seibold Barcelona, Barcelona, Spain Abbott Diabetes Care, Wiesbaden, Germany S. Del Prato Department of Clinical and Experimental Medicine, University of Pisa, Pisa, Italy Diabetes Ther the cohorts with TIR data reported an average The COVID-19 pandemic has highlighted clinically significant TIR improvement of 5% or the unmet need for wider application of more, possibly as a consequence of more accu- telemedicine and telemonitoring via CGM rate glucose monitoring and improved connec- for people with diabetes, including those tivity through telemedicine. Periodic with hyperglycaemia and early-stage consultations using telemedicine enables care diabetic ketoacidosis that may go of people with diabetes while limiting the need untreated during times of restricted for in-person attendance at diabetes clinics. clinical access. Reports that sustained hyperglycaemia and The efficacy of diabetes digital health early-stage diabetic ketoacidosis may go ecosystems has been validated during the untreated because of the lockdown and con- COVID-19 public health emergency and cerns about potential exposure to the risk of argues for accelerated implementation of infection argue for wider access to glucose tele- these models of care in diabetes. monitoring. Therefore, in this paper we have critically reviewed reports concerning use of telemonitoring in the acute hospitalized setting as well as during daily diabetes management. Furthermore, we discuss the indications and implications of adopting telemonitoring and INTRODUCTION telemedicine in the present challenging time, as well as their potential for the future. Following the outbreak and spreading of COVID-19, it became clear from analyses car- ried out in China, USA, Italy and UK that people Keywords: Diabetes; Insulin pumps; with diabetes hospitalized for COVID-19 have Continuous glucose monitoring; Ambulatory higher risks of more severe outcomes, including glucose profile; Virtual care; COVID-19; increased risk of death [1–5]. A meta-analysis Mortality; Risk management; Telemonitoring; including 83 studies [6] confirmed that people Telemedicine with pre-existing diabetes admitted to hospital because of COVID-19 had a twofold higher risk of requiring intensive unit care (n = 22 studies; Key Summary Points odds ratio 2.10) and a nearly threefold higher risk of in-hospital death (n = 15 studies; odds Despite restricted access to standard ratio 2.68). However, there are significant clinical care during the COVID-19 regional differences, with diabetes-related risks pandemic, glycaemic control has not being more prominent in non-Asian versus deteriorated for people with type 1 Asian countries [6]. Outcomes have been diabetes using telemonitoring of glucose claimed to differ between T1D and T2D with the data. risk of dying in hospital with COVID-19 being The increased risk profile for severe 3.5-fivefold higher in T1D and twofold in T2D COVID-19 disease on hospital admission compared to people without diabetes [5, 6]. for people with diabetes can be mitigated When adjusted for previous hospital admissions by application of telemonitoring and with coronary heart disease, cerebrovascular glucose-lowering treatment immediately disease or heart failure, these increased risks for following admission. dying in hospital after contracting COVID-19 are 2.86 and 1.81 in T1D and T2D, respectively. Routine telemonitoring can identify Mortality risk increases with age both in T1D groups of at-risk individuals with diabetes and T2D and is emphasised in people over who need in-person consultation and 70 years old, with the highest risks being for care, as well as those who may be people over 80 years of age (relative risk 4.79 in successfully managed with telemedicine. T1D, 4.52 in T2D) [7]. However, it appears, that Diabetes Ther in T1D, only people over 50 years old, with visual summary of the known risk factors is longer duration of the disease (80% with more shown in Fig. 1. than 15 years of disease) and raised HbA1c [ The need to reduce the risk of community 10% (86 mmol/mol) are at higher risk of severe transmission of COVID-19 for vulnerable clinical outcomes of COVID-19 [4]. Amongst groups led to national guidance for people with people with diabetes, those with prior hospital diabetes to undertake extreme social distancing admission for diabetic ketoacidosis (DKA) or and shielding in place that minimises infection severe hypoglycaemia in the past 5 years were risk, including attendance at diabetes clinics. three times more likely to develop fatal or crit- Extreme social distancing during the global ical care unit-treated COVID-19 [8]. Notably, COVID-19 pandemic meant that people with children and adolescents with diabetes do not diabetes have reduced or had no access to their appear to be at risk of more severe outcomes regular face-to-face diabetes services, being with COVID [9, 10]. Prevalence, outcomes and replaced in most cases with a reduced frequency prognosis in this group appear to be no different of telephone or video consultations. For people from their non-diabetic counterparts and con- with diabetes who can uplink glucose data to sistently milder than adults with diabetes. A the cloud and connect remotely with their healthcare provider, such as users of continuous Box 1. Definitions for Telemedicine and Telemonitoring in Context of Diabetes Care Telemedicine Managing diabetes care using telecommunications technology, including telephone and video conferencing, to deliver care at a distance. In this way, a healthcare professional (HCP) in one location can provide a medical consultation, including treatment adjustment, to a patient at a distant site. Telemedicine can also involve multiple HCPs in a synchronous patient consultation or care review. Telemonitoring The use of diabetes-specific medical devices that can monitor, transmit and share indicators of diabetes health between a person with diabetes and their diabetes care team. Telemonitoring allows care of patients at home or at other locations remote from their HCPs, using mobile phones, tablet computers, and desktop computers. Telemonitoring allows information, such as interstitial glucose readings from CGM systems, to be stored and reviewed by separate members of the diabetes care team, including the patient. Fig. 1 Risk matrix for people during the COVID-19 estimated glomerular filtration rate (ml/min/1.73 m2), pandemic. Summary of the known risk factors for T1D type 1 diabetes, T2D type 2 diabetes infection and disease severity with COVID 19. eGFR Diabetes Ther Table 1 Studies reporting glycaemic metrics before and during COVID-19 social lockdown Study Region Adults/ Type System Change in glycaemic children (n) metrics Bonora BM [17] Italy Adults (33) T1D FreeStyle Libre AG, mg/dL 177–160 (p = 0.005) %TIR 54–65.2 (p = 0.01) %TAR1 42.3–31.6 (p = 0.016) SD, mg/dL 58.9–53.2 (p \ 0.05) Capaldo B [19] Italy Adults (207) T1D FreeStyle Libre (n = 130) %TIR 55.6–58.2 Medtronic Guardian (n = 47) (p = 0.002) Dexcom G6 (n = 18) %TBR2 1.42–0.58 (p \ 0.001) Eversense (n = 12) %CV 35.9–34.7 (p = 0.001) Schiaffini R [21] Italy Children (22) T1D Dexcom G6 %TIR 61.45–66.41 (p \ 0.001) %TAR1 34.73–29.86 (p \ 0.001) TDD, IU/day 7.9–5.3 (p \ 0.001) Christoforidis A Greece Children (34) T1D Medtronic Enlite %CV 39.5–37.4 [77] (p = 0.011) Fernández E [78] Spain Adults (307) T1D FreeStyle Libre AG, mg/dL 167–158 (p \ 0.001) %TIR 57.8–62.5 (p \ 0.001) %TBR1 4.9–5.5 (p \ 0.001) %TAR1 37.3–32.0 (p \ 0.001) eA1c 7.4–7.1 (p \ 0.001) Mesa A [15] Spain Adults (92) T1D Free Style Libre (n = 75) %TIR 59.3–62.6 (p \ 0.05) Dexcom G5 (n = 17) %TAR1 34.4–30.7 (p \ 0.05) GMI, % 7.2–7.0 (p \ 0.05) Diabetes Ther Table 1 continued Study Region Adults/ Type System Change in glycaemic children (n) metrics Cotovad-Bellas L Spain Adults (44) T1D FreeStyle Libre No change during [79] lockdown Caruso I [80] Italy Adults (48) T1D FreeStyle Libre %TBR1 6.3–4.5 (p \ 0.001) %CV 38.3–35.1 (p = 0.008) Aragona M [81] Italy Adults (63) T1D FreeStyle Libre (n = 52) AG, mg/dL 165–158 Dexcom G6 (n = 11) (p = 0.04) %TIR 58–61 (p \ 0.05) %TAR1 38–34 (p \ 0.04) GMI, % 7.2–7.0 (p = 0.016) Dover AR [22] UK Adults (572) T1D FreeStyle Libre AG, mmol/L 9.6–9.3 (p \ 0.001) %TIR 53–56 (p \ 0.001) %TBR 3-3 (p = 0.005) %TAR1 42–39 (p \ 0.001) eA1c 61–58 (p \ 0.001) Di Dalmazi G [82] Italy Children T1D Freestyle Libre (n = 2) SD 67.4–64.3 (p = 0.029) B 12 years Dexcom G5 or G6 (n = 28) (30) Di Dalmazi G [82] Italy Teenagers T1D Freestyle Libre (n = 2) No change during 13–17 years Dexcom G5 or G6 (n = 22) lockdown (24) Diabetes Ther Table 1 continued Study Region Adults/ Type System Change in glycaemic children (n) metrics Di Dalmazi G [82] Italy Adults (76) T1D Freestyle Libre (n = 41) AG, mg/dL 164–160 Dexcom G5/G6 (n = 35) (p = 0.001) %TIR 57.7–61.3 (p \ 0.001) %TBR1 2.5–3.0 (p = 0.019) %TAR1 26.5–25.1 (p = 0.001) %TAR2 11.1–7.8 (p = 0.007) SD 63.0–59.6 (p = 0.016) eA1c, % 7.2–6.8 (p \ 0.001) GMI,% 7.7–7.6 (p = 0.001) Brener A [83] Israel Children (102) T1D Dexcom G5 No change during lockdown Pla B [84] Spain Adults (50) T1D FreeStyle Libre AG, mg/dL 160.3–150.0 (p \ 0.001) %TIR 57.5–65.8 (p \ 0.001) %CV 40.7–36.4 (p \ 0.001) eA1c, % 7.2–6.8 (p \ 0.001) GMI,% 7.2–6.9 (p \ 0.001) Prabhu Navis J UK Adults (269) T1D Freestyle Libre (n = 190) %TIR 57.5–59.6 [18] Dexcom G6 (n = 79) (p = 0.002) %CV 37–36 (p = 0.003) Barchetta I [14] Italy Adults (50) T1D FreeStyle Libre and other CGM, AG, mg/dL 154–165 unspecified (p = 0.027) %TIR 75–69 (p \ 0.001) %TBR1 6–10 (p = 0.002) eA1c 7.3–7.5 (p = 0.031) Diabetes Ther Table 1 continued Study Region Adults/ Type System Change in glycaemic children (n) metrics Predieri B [85] Italy Children (62) T1D Dexcom G6 %TIR 60.0–62.1 (p = 0.008) %TAR1 37.8–35.7 (p = 0.048) %TBR1 2.63–2.13 (p = 0.001) %TBR2 0.5–0.34 (p = 0.002) Braune K [23] Germany Children (28) T1D Freestyle Libre (n = 9) %TIR 46.9–56.3 Dexcom G5/G6 (n = 6) (p = 0.016) Medtronic Guardian (n = 13) %TAR1 21.3–13.9 (p = 0.013) Potier L [20] France Adults (1378) T1D FreeStyle Libre AG, mmol/L 9.1–8.7 (p \ 0.001) Boscari F [24] Italy Adults (79) T1D FreeStyle Libre (n = 52) AG, mg/dL 161.1–156.3 Dexcom (n = 16) (p = 0.001) Other CGM (n = 3) %TIR 63.6–66.3 (p \ 0.001) %TAR1 33.4–30.5 (p = 0.002) Sánchez Conejero Spain Children (80) T1D FreeStyle Libre (n = 15) %TIR 72.1–74.8 (p = 0.01) M [86] Medtronic Guardian (n = 52) %TAR2 4.6–3.7 Dexcom G6 (n = 13) (p = 0.038) %TBR1 4.6–3.2 (p \ 0.001) %TBR2 1.2–0.7 (p \ 0.001) %CV 35.8–33.1 (p \ 0.001) Diabetes Ther Table 1 continued Study Region Adults/ Type System Change in glycaemic children (n) metrics Rachmiel M [25] Israel Children (195) T1D Unspecified CGM AG, mg/dL 164–160 (p \ 0.001) %TIR 59–63 (p \ 0.001) %TAR1 32.0–38.6 (p = 0.005) %CV 37.1–35.6 (p = 0.002) eA1c, % 7.35–7.20 (p \ 0.001) Marigliano M [87] Italy Children (233) T1D FreeStyle Libre (n = 98) AG, mg/dL 178.6–169.1 Dexcom G6 (n = 100) (p \ 0.001) SMBG (n = 35) %TIR 52.6–58.0 (p \ 0.001) %TAR1 43.4–38.0 (p \ 0.001) SD 68.0–63.6 (p \ 0.001) GMI, % 7.60–7.37 (p \ 0.001) Alharthi SK [26] Saudi Adults (101) T1D Unspecified CGM AG, mg/dL 180–159 Arabia (p \ 0.001) %TIR 46–55 (p \ 0.001) %TAR1 48–35 (p \ 0.001) GMI,% 7.7–7.2 (p = 0.03) van der Linden J USA NA (65,067) All Dexcom G6 %TIR 59–61 (p \ 0.001) [88] Diabetes Ther Table 1 continued Study Region Adults/ Type System Change in glycaemic children (n) metrics Viñals C [16] Spain Adults (59) T1D Medtronic 640G ? Enlite AG, mg/dL 153.2–147.8 (p \ 0.001) %TIR 67.6–69.8 (p \ 0.017) eA1c, % 6.94–6.75 (p \ 0.001) Studies are listed in date order as disclosed on PubMed. Significant changes in glycaemic metrics from before to during COVID-19 lockdown are presented. Each number sequence indicates change from before to during lockdown with p value AG average glucose, eA1c estimated HbA1c, %CV % coefficient of mean glucose, GMI glucose management indicator, %TIR percentage time in range 70–180 mg/dL (3.9–10 mmol/L), %TBR1 percentage time below range \ 70 mg/dL (3.9 mmol/ L), %TBR2 percentage time below range \ 54 mg/dL (3.0 mmol/L), %TAR1 percentage time above range [ 180 mg/dL (10 mmol/L), %TAR1 percentage time above range [ 250 mg/dL (13.8 mmol/L), TDD total daily dose of bolus insulin, NA no information about demogaphics glucose monitoring (CGM), including flash that has enabled clinicians and people with glucose monitoring, clinical review of glucose diabetes to view and discuss glucose downloads control and shared decision-making in regard to together during this period. In the absence of treatment adjustment has been accomplished routine laboratory HbA1c testing, several stud- via telemedicine and glucose telemonitoring ies, since the start of the COVID-19 pandemic, tools. Substantial evidence now indicates that have used percentage time in range (%TIR) and this has been effective (Table 1). Routine care other CGM-derived metrics [12] to determine was then effectively switched to telemedicine the degree of glycaemic control. %TIR is indeed and digital services, including for diabetes becoming established as a significant indicator reviews and self-management support, and of immediate and medium-term glucose control published data has demonstrated that glucose [13]. Table 1 summarises the various studies control was not impaired while access to regular published between March 2020 and March diabetes clinical services was interrupted [11]. 2021. Each of the studies selected individual This review summarises available data on the before and during lockdown criteria for com- importance both of telemonitoring and tele- parison of glucose metrics, typically bench- medicine, as defined in Box 1, in effective dia- marking lockdown as starting in the middle to betes management for reducing the risks for end of March 2020. COVID-19 infection. This article is based on previously conducted studies and does not contain any new studies with human partici- TIME IN RANGE DURING COVID-19 pants or animals performed by any of the authors. Across the 27 cohorts studied, %TIR increased significantly in 19 of them, and did not change in 7 of them. Significantly, time in range (TIR) STUDIES REPORTING GLUCOSE 70–180 mg/dL/3.9–10 mmol/L improved across METRICS BEFORE AND DURING 18/27 cohorts with a median change (range) of COVID-19 SOCIAL LOCKDOWN 3.3% (- 6.0 to 11.2%), covering 67,587 indi- viduals using CGM sensors. Notably, 30% of the cohorts with TIR data reported clinically signifi- Studies were reviewed that reported data on cant TIR improvements of 5% or more [13]. Only remote monitoring with CGM technologies Diabetes Ther in the study by Barchetta et al. [14] did %TIR is interrupted has been shown by another Italian deteriorate during the lockdown period. In this study on 22 children [21] that showed a signifi- study, lockdown-associated stress due to loss of cant increase in %TIR and reduction in %TAR work or furlough from work independently pre- during COVID-enforced home care for this dicted impaired TIR after adjustment for poten- group. Notably, this improvement seems to be tial confounders. Subjects in this study also associated with greater use of correction boluses. experienced an increase in percentage time The downside of these otherwise encouraging above range (%TAR) and percentage time below assessments was highlighted in the study by range (%TBR) during the lockdown. This aligns Dover and colleagues [22]. Although CGM met- with observations from Bonora et al. [17], who rics improved across the study group (n = 572), have reported that improvements in %TIR and they also demonstrated a deterioration of at least %TAR were restricted to those who stayed at 5% TIR in a subset of patients (n = 125) and that home during lockdown, whereas those still this was significantly associated with higher experiencing the stress of on-site working saw no levels of socio-economic deprivation. This must change. Reductions in TAR and in average glu- be included in learnings from the enforced cose are present in 13/27 cohorts and 13/27 shielding during COVID-19 in order to ensure cohorts, respectively, indicative of the increases that a care gap does not emerge in the applica- in %TIR. %TBR \ 70 mg/dL (\ 3.9 mmol/L) tion of diabetes technology once the COVID-19 decreased only in 4/27 cohorts and increased in pandemic is over. 3/27 cohorts, with no change in the rest, Most of the studies included did not include including in subjects prone to hypoglycaemia a formal telemedicine component, rather they [16, 17]. When reported, improvements in %TIR collected data from an observational standpoint in T1D after lockdown are negatively associated and made no assessment of the nature of tele- with baseline %TIR [18], with greater improve- medicine contacts between patients and their ments seen for people with lower %TIR pre- diabetes HCPs, although these are acknowl- lockdown. Baseline HbA1c is also predictive of edged to have happened. However, 4/27 studies greater improvement in %TIR, such that people did involve an active telemedicine intervention in whom %TIR improved by 5% or more were and in each of these studies [23–26] there was more likely to have a higher pre-lockdown an improvement in glucose metrics. Most tell- HbA1c [18]. ingly, in the study by Alharthi [26], CGM users It has been suggested that restrictions on were categorized into two groups, one that community, social and leisure time provided participated in a single telemedicine consult more time for diabetes self-management during during the 6-week lockdown study period the lockdown periods and may have contributed (n = 61) and one that did not (n = 40). Those to improved glycaemic control in this period. who attended a telemedicine visit had a signif- Again, in favour of this idea are the outcomes icant improvement in average glucose, GMI, from the cohort in Italy, showing that CGM %TIR and %TAR. In contrast, there were no users who stayed at home during lockdown had significant changes in any CGM metrics in the significant improvements in average glucose and group that did not receive a telemedicine %TIR, whereas those who continued working consultation. did not. Capaldo and colleagues [19] indicated that improvements in %TIR and reduced gly- caemic variability amongst their cohort are DIABETES CONTROL IN ABSENCE associated with reduced physical activity, more OF TELEMONITORING regular meal patterns and a change in sleeping patterns. Equally, the reduction in average glu- The question must be asked as to the glucose cose reported by Potier [20] was associated with control of people with diabetes who have not positive behavioural changes. The importance had access to telemonitoring with CGM. Over- and efficacy of greater parental care of children all, the expected deterioration in glycaemic with T1D while access to standard paediatric care control is only variably evident. Two studies Diabetes Ther have indicated that people with T2D have seen target \ 7.5% (58 mmol/mol) is associated with their HbA1c rise during COVID [27, 28] but in at a cost of €28,997, whereas a person with T1D least one of these studies the rise has been for and higher HbA1c (C 7.5%; 58 mmol/mol) people with diabetes less than 50 years of age would cost €57,244. For people with T2D these [27] but not older. Balancing these observations costs are €25,018 and €46,130. The lesson is that are studies showing that COVID-19 home con- maintaining good glucose control is of high finement did not negatively impact glucose value, even if telemonitoring and telemedicine control in patients with T2D [29], although have not prevented a COVID-related older age and insulin therapy are associated hospitalization. with greater risk of deterioration of glucose control, and a separate study has observed a fall in HbA1c during the COVID restrictions [30]. A HYPERGLYCAEMIA IS study from the Netherlands found that, during A SIGNIFICANT RISK FACTOR the COVID-19 lockdown, people with relatively FOR COVID INFECTION, well-controlled T1D or T2D showed increased HOSPITALIZATION AND DEATH stress and weight gain but no deterioration in glucose control [31]. A section of the study That diabetes is a significant risk factor for group was CGM users so it is hard to interpret COVID infection and subsequent severe disease these findings. In T1D, an increase both in is accepted [4, 6, 35]. A significant part of this average glucose and HbA1c was reported in a risk profile can be attributed to hyperglycaemia. small cohort of people in India [32]; however, Amongst a cohort of 183 people with diabetes this appears to be mainly due to non-availabil- in Israel, HbA1c was shown to predict con- ity of insulin or self-monitoring blood glucose tracting COVID-19 and subsequent hospitaliza- (SMBG) strips during the lockdown period. tion [33] (Fig. 2). Amongst this cohort, those with an HbA1c of 8.0–8.9% (64–74 mmol/mol) had a 1.56 increased odds ratio contracting COSTS OF COVID IN DIABETES COVID compared to people with HbA1c \ 7.0% POINT TO NEED FOR EFFECTIVE (\ 53 mmol/mol), and those with HbA1c TELEMEDICINE C 9.0% (C 75 mmol/mol) had an odds ratio of 1.84. Regarding hospitalization following For healthcare services, a move towards tele- infection, compared to an HbA1c \ 7.0% monitoring for care of people with chronic (\ 53 mmol/mol), levels of 7.0–7.9% (53- conditions, including diabetes, has been part of 63 mmol/mol), 8.0–8.9% (64–74 mmol/mol) a long-term plan [33]. However, the COVID-19 or C 9.0% (C 75 mmol/mol) increased the odds public health emergency has forced the accel- of hospitalization by 2.4, 3.2 and 3.4, respec- eration of this process. The increased risks of tively [36]. severe disease and death for people with dia- Regarding mortality from COVID in hospi- betes contracting COVID are evident but these tal, the OpenSAFELY analysis of over 17 million are also matched by a significant financial primary care records [37] indicates that, when impact. An analysis of hospital admissions adjusted for all covariables, adults with diabetes across Europe [34] during the period Jan- and HbA1c of \ 7.5% (\ 58 mmol/mol) have a uary–June 2020 has calculated that the total 1.31 hazard ratio for death with COVID, which costs for treating people with diabetes admitted rises to 1.95 for HbA1c C 7.5% to hospital with COVID-19 during this period (C 58 mmol/mol). More specifically, Holman was €3.38 billion, which is approximately 24% and colleagues [4] showed that people with T1D of all COVID-related admission costs. Of par- and HbA1c C 10% (86 mmol/mol) had a 2.23- ticular note is that varying costs are associated fold increased risk of death compared to those with the type of diabetes and the level of with an HbA1c of 6.5–7.0% (48–53 mmol/mol). metabolic control [34]. Thus, each admission People with T2D had a 1.22-fold risk of death for a person with T1D who has HbA1c close to Diabetes Ther Fig. 2 Hyperglycaemic risk profiles for COVID-related hazard ratio [37] or odds ratio [36] assessments. Risk } disease and mortality. For HbA1c \ 7.5% factors categorized by separate diabetes type indicated on (58 mmol/mol) [37]. *For HbA1c C 7.6% the bars only where known (60 mmol/mol) [4]. Risk profiles compared to HbA1c \ 7.0% (53 mmol/mol). Data derived from relative risk [4], with HbA1c levels C 7.6% (59 mmol/mol) [44] showed that %TAR [ 160 mg/dL compared to those with HbA1c of 6.5–7.0% ([ 8.9 mmol/L) or %TBR \ 70 mg/dL (48–53 mmol/mol), rising to 1.36 above 9.0% (\ 3.9 mmol/L) is associated with up to 2.45- (75 mmol/mol) and 1.61 above 10% fold increased odds ratio of ICU admission, (85 mmol/mol). These significant associations mechanical ventilation or hypotension/multi- reinforce the evidence that hyperglycaemia at ple organ dysfunction. Length of hospital stay admission increases in-patient mortality with was also associated with %TAR and %TBR. Sig- COVID-19, both for people with pre-existing nificantly, the odds ratio for adverse outcomes diabetes or without [38–40]. The importance of was also increased with glycaemic variability hyperglycaemia in COVID infection is linked to (GV) above a coefficient of variation (CV) of a number of hypotheses [41–43] whereby SARS- 36%. The implication is that maintaining TIR CoV-2 infection is facilitated then potentiated 70–160 mg/dL (3.9–8.9 mmol/L) as close to 70% by the generalized multisystemic inflammation as possible is a management strategy that process observed that further drives the hyper- should be targeted for all patients admitted to glycaemic state. hospital with hyperglycaemia and COVID. The need to maintain good glucose control is Importance of Telemonitoring Following supported by the analysis of Klonoff and col- COVID-Related Admission leagues [45], who noted that severe hypergly- and Hyperglycaemia caemia ([ 250 mg/dL/[ 13.9 mmol/L) on days 2–3 after admission was independently Glucose telemonitoring is emphasised for associated with high mortality (adjusted hazard patients admitted to hospital with COVID, both ratio 7.17; 95% CI 2.62–19.62) compared with those with diabetes and those without diabetes patients with glucose \ 140 mg/dL (\ 7.8 mmol/ but high admission glucose. Using CGM in 35 L). Notably, reducing high admission glucose to patients with COVID and diabetes, Shen et al. 141–180 mg/dL (7.8–10 mmol/L) by day 2–3 was shown to significantly reduce the risk of Diabetes Ther mortality [45]. The impact of concurrent COVID been linked to more severe organ damage with DKA and hyperglycaemic hyperosmolar amongst people with newly diagnosed diabetes syndrome (HSS) on post-admission outcomes [51], suggestive of a metabolic disruption that is [46] also argues for the application of immediate mitigated following diabetes therapy. telemonitoring and glucose-lowering treatment In these studies, the designation of ‘newly following admission. Given that regular glucose diagnosed’ refers to their hyperglycaemia state monitoring using standard point-of-care testing detected in hospital rather than a ‘recent diag- exposes hospital teams to risk of COVID infec- nosis’. However, the increased risk is not solely tion, the importance of telemonitoring that aids due to hyperglycaemia, since their glucose anti-glycaemic therapy whilst reducing frequent levels are typically lower than people with pre- patient contact is strongly indicated. existing diabetes or without diabetes at the point of admission [2]. Of course it is difficult to Risk of New-Onset Diabetes Because ascertain if a proportion of these patients may of COVID-19 Infection have had undiagnosed diabetes at the point of infection. Therefore it is important to better Evidence that infection with COVID-19 may understand the metabolic fragility of a person precipitate diabetes is also emerging [42]. The with diabetes at the point of diagnosis and SARS-CoV virus, which caused the SARS epi- immediately thereafter, and also to take steps to demic, is reported to bind angiotensin-con- overcome this state. Further research is required verting enzyme 2 (ACE2) receptors in the to establish the extent and duration of ‘new pancreas with consequent damage to islet cells onset’ risk following diagnosis and the patho- and reduced insulin release [40, 47]. Compared physiology associated with it. This would be an to healthy siblings, SARS infection resulted in important role for telemonitoring with CGM symptomatic T2D during hospitalization in half systems, which can both help educate the new of cases, with 5% remaining diabetic 3 years patient and also help improve their glycaemic later. During the COVID-19 pandemic, SARS- dysregulation. CoV-2 infection has been associated with the subsequent onset of insulin-dependent diabetes Managing New-Onset Type 1 Diabetes [48], presenting with DKA in a young man with Using Telemedicine severe loss of b-cell function but no islet-specific autoantibodies. Direct SARS-CoV-2-mediated b- Insulin initiation in new-onset T1D is another cell damage is proposed as the cause of the T1D important aspect of telemedicine for people manifested in this case [48]. Similar associations with diabetes that has typically emphasised in- have been made for children [49]. person clinic attendance and face-to-face train- ing. This standard approach has been another casualty of the COVID-19 pandemic, despite the NEWLY DIAGNOSED DIABETES evidence that this group are at high risk of AND NEED FOR EARLY GLUCOSE COVID-related harm. In two case reports, tele- CONTROL medicine has been used effectively and safely as part of the insulin initiation, including dose In addition, several strands of evidence indicate titration, as well as to provide education on that newly diagnosed diabetes should be a clear diabetes management and support. These target for immediate glucose control. In the reports describe a newly diagnosed adult and a context of COVID-19, people with newly diag- 12-month-old child [52]. Both cases did involve nosed diabetes are at most risk of COVID-re- initial emergency hospital admission and lated admission to intensive care or death [50], assessment, and both patients were started on with three times the risk of people with pre- either multiple daily injections of insulin (MDI, existing diabetes [2, 50] and an increased length adult subject) or on continuous subcutaneous of stay in hospital. This increased mortality has insulin infusion (subcutaneous insulin infusion Diabetes Ther (CSII), paediatric subject) as part of in-hospital neurology assessment within the therapeutic care. However, follow-up glycaemic monitor- window as a consequence of anxieties over ing, insulin adjustment and education were COVID transmission. substantially provided through telemedicine Fear of attending hospitals since March 2020 and using ambulatory glucose profiles and has had significant impacts for people with interpretive summary reports generated diabetes, with consequent outcomes for non- through CGM initiated while in hospital. In COVID severe illness [57]. A survey of 4075 both cases, the subjects were successfully man- paediatricians in the UK and Ireland [58] indi- aged via telemedicine. In the UK, led by the cated that symptomatic DKA was the most national Diabetes Technology Network, a common problem that resulted from fear of donation of sensors was used to support rapid attendance, separately from new-onset diabetes. discharge and remote follow-up of patients in Reports indicate that during the COVID-19 this setting [53]. pandemic lockdown, the rate of DKA has increased amongst children with newly diag- nosed T1D [59], including an increase in more FEAR OF INFECTION CHANGES severe cases [59–61]. This is accompanied by ENGAGEMENT WITH HEALTH reports of more prevalent presentation of severe SERVICES FOR PEOPLE DKA in COVID-positive children with estab- WITH DIABETES lished T1D [61]. It is likely that the increase in severity in new or established cases coincides People with diabetes have significant COVID- with delayed attendance at hospitals due to fear 19-specific worries related to their diabetes. of COVID. Regular education of the whole Social distancing gives rise to stressors that lead society about the symptoms of diabetes could to anxiety and this may be exacerbated for contribute to faster diagnosis of T1D and people with diabetes who can be subject to reduction of DKA prevalence, along with more advice for stricter isolation. A survey of 1396 rigorous adherence to ‘sick-day rules’ which are people with diabetes in Denmark [54] found recommended for established T1D in such sit- that the most prevalent diabetes-specific worry uations [62, 63]. related to COVID-19 was fear of being overly Risk of DKA for people with T2D may be affected by the virus (56% respondents) along exacerbated by COVID infection. A series of with being labelled as a high-risk group (39% cases have been described in which people with respondents) and an inability to manage their T2D and no history of DKA have presented at diabetes if infected (28% respondents). In gen- hospital with new-onset DKA, likely precipi- eral, improvements in psychosocial health tated by COVID infection [64]. People with T1D started to occur during the first reopening phase are also at highest risk of DKA during COVID- after the initial survey responses. related social lockdown, possibly related to It is likely that this fear of infection con- potential changes in access to insulin or SMBG tributes to a reluctance to visit medical centres test strips [64]. Under these circumstances the during COVID. There is evidence that rates of value of telemonitoring with CGM can make admission for significant macrovascular disease the difference between early intervention and are lower than usual, which may reflect reluc- treatment, or a potentially life-threatening tance of people to attend hospital. Although acute DKA episode. Both the FUTURE study in not specific to diabetes, data from Kaiser Per- three specialist diabetes centres in Belgium [65] manente in the USA showed a 48% reduction in and the larger nationwide RELIEF study in hospitalizations for acute myocardial infarction France [66] have reported a significant fall in during the early weeks of lockdown [55]. Simi- acute diabetes events (ADE), including DKA, larly, a study in France [56] has indicated that, after reimbursement of the FreeStyle Libre CGM although the overall incidence of stroke system was approved within each health ser- remained the same, fewer patients presented for vice. A recent retrospective, real-world analysis of a medical insurance claims database in the Diabetes Ther USA has shown a similar fall in ADE and related hospitalization and death [6, 35, 36]. Impor- hospitalizations, including for DKA, once the tantly, these risks reflect known attributes of FreeStyle Libre system was prescribed for people people with diabetes, including type and dura- with diabetes [67]. Given the additional tion of diabetes, longer-term control (HbA1c), COVID-related risks for people with diabetes short-term hyperglycaemia, comorbid and and prior hospitalization for DKA [8], the ben- intercurrent disease, ethnicity and socio-eco- efits of telemonitoring are strongly emphasised nomic status. Practical guidance for managing as they enable diabetes care at a distance and people with diabetes during COVID-19 has been minimise metabolic deterioration and ADEs. published [62] and telemonitoring with CGM means that important parts of the risk profiles can be managed, both via telemedicine and RISK STRATIFICATION through in-person consultation at a diabetes AND PRIORITIES centre. The future of telemedicine is not simply FOR TELEMONITORING, about keeping people away from hospitals, it is TELEMEDICINE OR ATTENDANCE equally about knowing who should be asked to come to the clinic and when. Thus, we see a IN CLINIC future for diabetes care that prioritises people with diabetes for appropriate consultations based The current reliance on telemedicine for on their glucose metrics and other risk markers. managing people with diabetes will persist, not Naturally, further adoption of telemedicine is just in the near future but as part of standard care part of clinical care for the many people whose as diabetes services evolve to accommodate the diabetes can be managed with a clear view of successes and challenges of the COVID pan- their glucose metrics. It must also be pointed out demic. However, it is important to acknowledge that certain care processes will require in-clinic that telemedicine must be viewed in context. visits at a certain frequency, e.g. diabetes-related Although telemedicine during COVID has been foot disease. Similarly, people with more acute largely welcomed as a positive experience in disease will not easily fit the transition to tele- people with T1D, it is not universally appreci- medicine. These include those with significant ated. A survey of 7477 individuals across Europe, hypoglycaemia unawareness or fear of hypogly- North America, South American, Africa and Asia caemia, as well as people with frequent diabetes- has highlighted both the pros and cons [68]. related admission for DKA or hyperosmolar Thirty per cent of respondents reported that hyperglycaemic state (HHS) [69]. This also their healthcare access had suffered during applies to people with diabetes with comorbid COVID and the majority (86%) found remote disease and risks for cardiovascular or cere- consultations useful, with 75% indicating that brovascular events, who should be seen in-clinic they would continue with remote appointments at a frequency that reflects the urgency of their in the future. This covered all age groups, condition. including those over 65 years old. However, a group of respondents reported a negative per- ception of telemedicine. These were typically DIABETES SERVICES ALREADY male and poorly controlled (HbA1c [ 9.0%). The EMPLOY TOOLS NEEDED context of their reluctance to participate in tel- emedicine was unclear, whether it was down to FOR TELEMEDICINE lack of engagement with their own diabetes or AND TELEMONITORING lack of engagement with telemedicine. What is clear is that telemedicine without The virtual diabetes clinic has been a goal of telemonitoring in diabetes is a harder proposi- healthcare systems and expert diabetes profes- tion. People with diabetes are at greater risk at sionals for some time. Based around the grow- every step of the natural history of COVID-19, ing number of technologies that enable diabetes from first infection through to severe disease, care, telemedicine has been proposed as an Diabetes Ther important solution to the need to expand care the group using CGM alone, compared to those for the benefit of people with diabetes, whilst using insulin injections and SMBG. This would improving efficiencies and rationalizing costs. suggest that the full impact of each of these As recently as February 2020, less than a month enabling technologies, though demonstrated, before the COVID pandemic was declared, an can be further improved. expert panel endorsed the opportunities and Together, these real-world insights indicate challenges of making telemedicine a standard of that the need for technology interoperability is care for the management of diabetes [33]. In already being met for effective telemedicine in doing so they identified barriers for system diabetes, with new ways of working that lever- interoperability within a diabetes health age existing digital health ecosystems. It is clear ecosystem and HCP reimbursement. In the UK, that this can improve in the future, with sys- an expert panel recommended a risk stratifica- tematic application of additional tools that can tion and risk-based follow-up approach using enhance the emerging practice of telemedicine, telemonitoring to help stratify people into red, but it is effective now. A further barrier to tele- amber or green categories [69]. medicine is also being dismantled as a conse- The COVID-19 pandemic has shortcut some quence of COVID, that of reimbursement. of these important conversations, disrupting Healthcare payers are widening existing provi- the system and driving change in a manner that sions for telemedicine or creating them where could not have been predicted. In doing so, the they did not exist [74–76]. most agile services have emphasised and adop- ted systems and technologies that support people with diabetes in new models of care that CONCLUSIONS minimise their need to attend busy medical centres. It is not a coincidence that the early Telemedicine and telemonitoring have been adopters of new service designs are paediatric transformed from an aspirational goal to diabetes clinics [9, 23, 70], one of which reports become the de facto standard of care for dia- more than 80% of scheduled visits had been betes management during the COVID-19 public accomplished by telemedicine within 3 weeks health emergency in the developed world. This of starting, with fewer missed appointments has been in response to the need to protect [70]. people with diabetes who are at immediate and It can be argued that paediatric services are greater risk of severe disease, hospitalization more agile in this regard because diabetes and death from COVID-19. However, we must management enabling technology is more acknowledge that COVID is also an instructive prevalent amongst younger children with dia- use case that highlights the vulnerability of betes [71]. The value of this is confirmed by people with diabetes in situations that disrupt recent outcomes from the SWEET project, a the process of standard care. Ultimately, as well multinational network of more than 100 pae- as managing people with diabetes at a distance, diatric diabetes centres with a mission to the value of routine telemonitoring must also establish standards of care in paediatric diabetes be to identify groups of at-risk individuals who worldwide [72]. Data on 25,654 children and need in-person consultation and care. adolescents from SWEET [73] showed that Telemonitoring of glucose data has been an children and adolescents using continuous CSII invaluable tool in maintaining glycaemic con- pumps, CGM systems or both had significantly trol for a significant subset of those able to lower HbA1c compared to those using insulin access this technology, most evidently for peo- injections and SMBG. The proportion of DKA ple with T1D, including paediatric and adult episodes was lower in the group using an insulin groups. The case for wider application of glu- pump alone or in tandem with a CGM sensor. cose telemonitoring is clear but still faces sig- Of note, reported episodes of severe hypogly- nificant hurdles. Central to these is caemia were lower in the group using an insulin acknowledgement of the value of telemonitor- pump without a CGM sensor, but increased in ing and reimbursement for this along with Diabetes Ther increased access to CGM systems. The Editors (ICMJE) criteria for authorship for this cost–benefit calculation has been strongly article, take responsibility for the integrity of weighted towards the benefits during COVID the work as a whole, and have given their but further action is needed to conclusively approval for this version to be published. demonstrate the long-term financial savings. Education in regard of telemonitoring and tel- Author Contributions. Thomas Danne and emedicine is also critical to the successful Alexander Seibold had the idea for the article, implementation of technologies that better all authors contributed to the literature search enable diabetes care, both for HCPs and for and data analysis. Thomas Danne, Catarina people with diabetes. Just as important are the Limbert and Alexander Seibold drafted the first imperatives to manage data for enhanced dia- version of the manuscript, all authors con- betes care. The COVID public health emergency tributed to subsequent versions and read and has shown what can be achieved when the approved the final manuscript. Editorial assis- current technology is put under the spotlight, tance in the preparation of this article was but this can be better supported with systematic provided by Rob Brines and Wiebke Jessen. changes to diabetes care. Integration of data Support for this assistance was funded by uplink and data sharing systems is a recognised Abbott Diabetes Care. unmet need in this context. Additionally, with better systems integration comes the require- Disclosures. Thomas Danne reports grants ment for an information technology (IT) spe- and personal fees from Abbott, AstraZeneca, cialist contribution, not as a support service but Boehringer, DexCom, Lilly, Medtronic, Novo as an established member of the diabetes mul- Nordisk, Roche, Sanofi outside the submitted tidisciplinary team. Action on these aspects of work and is a shareholder of DreaMed Ltd. diabetes digital health is a clear imperative. Catarina Limbert reports grants and personal Telemedicine is an acknowledged part of the fees from Abbott, Ipsen and Sanofi. Pratik post-COVID world but telemonitoring via CGM Choudhary reports grants and personal fees is still possible only for a subset of people with from Abbott, Dexcom, Lilly, Novo Nordisk, diabetes. The diabetes services adapting most Sanofi and Insulet. Manuel Puig-Domingo quickly to the needs of care of people with reports grants, consultant and speaker fees from diabetes appear to be those with the most Sanofi, Novo Nordisk, AstraZeneca, Lilly, experience and confidence of working with Recordati, Pfizer, Ipsen and Novartis outside the diabetes health technologies and creating their submitted work. Eric Renard reports personal own diabetes health ecosystems. These are often fees as consultant/advisor for Abbott, Air Liq- paediatric services. uide, AstraZeneca, Boehringer Ingelheim, Cell- novo, Dexcom, Eli Lilly, Insulet, Johnson & Johnson (Animas, LifeScan), Medirio, Med- tronic, Novo Nordisk, Roche Diagnostics, ACKNOWLEDGEMENTS Sanofi-Aventis and Tandem, and research grant/material support from Abbott, Dexcom, Insulet, Roche Diagnostics and Tandem. Ste- Funding. Sponsorship for this study and fano Del Prato has served on the scientific board journal’s Rapid Service Fee were funded by and received honoraria for consulting fees from Abbott Diabetes Care. AstraZeneca, Boehringer Ingelheim, Eli Lilly, Medical Writing Assistance. Editorial assis- GlaxoSmithKline, Merck & Co., Novartis Phar- tance in the preparation of this manuscript was maceuticals, Novo Nordisk, sanofi, servier, provided by Dr Robert Brines of Bite Medical Takeda Pharmaceuticals. Consulting. Compliance with Ethics Guidelines. This Authorship. All named authors meet the article is based on previously conducted studies International Committee of Medical Journal and does not contain any new studies with Diabetes Ther human participants or animals performed by 5. Istituto Superiore di Sanità. Report of characteristics any of the authors. of patients who died positive for COVID-19 in Italy. 2020 (in Italian). https://www.epicentro.iss.it/ coronavirus/bollettino/Report-COVID-2019_17_ Data Availability. Data sharing is not marzo-v2.pdf. Accessed 18 June 2021. applicable to this article as no new data sets were generated during the current study. 6. Mantovani A, Byrne CD, Zheng M-H, Targher G. Diabetes as a risk factor for greater COVID-19 severity and in-hospital death: a meta-analysis of Open Access. This article is licensed under a observational studies. Nutrition Metab Cardiovasc Creative Commons Attribution-NonCommer- Dis. 2020;30:1236–48. cial 4.0 International License, which permits any non-commercial use, sharing, adaptation, 7. Barron E, Bakhai C, Kar P, et al. Associations of type 1 and type 2 diabetes with COVID-19-related distribution and reproduction in any medium mortality in England: a whole-population study. or format, as long as you give appropriate credit Lancet Diabetes Endocrinol. 2020;8:813–22. to the original author(s) and the source, provide a link to the Creative Commons licence, and 8. McGurnaghan SJ, Weir A, Bishop J, et al. Risks of and risk factors for COVID-19 disease in people indicate if changes were made. The images or with diabetes: a cohort study of the total popula- other third party material in this article are tion of Scotland. Lancet Diabetes Endocrinol. included in the article’s Creative Commons 2021;9:82–93. licence, unless indicated otherwise in a credit 9. Danne T, Limbert C. COVID-19 pandemic, type 1 line to the material. If material is not included diabetes and technology: why paediatric patients in the article’s Creative Commons licence and are leading the way. Lancet Diabetes Endocrinol. your intended use is not permitted by statutory 2020;8:465–7. https://doi.org/10.1016/S2213- regulation or exceeds the permitted use, you 8587[20]30155-8. will need to obtain permission directly from the 10. Cardona-Hernandez R, Cherubini V, Iafusco D, copyright holder. To view a copy of this licence, Schiaffini R, Luo X, Maahs DM. Children and youth visit http://creativecommons.org/licenses/by- with diabetes are not at increased risk for hospital- nc/4.0/. ization due to COVID-19. Pediatr Diabetes. 2021;22:202–6. 11. Wake DJ, Gibb FW, Kar P, et al. Endocrinology in the time of COVID-19: remodelling diabetes ser- REFERENCES vices and emerging innovation. Eur J Endocrinol. 2020;183:G67-77. 1. Zhou F, Yu T, Du R, et al. Clinical course and risk 12. Danne T, Nimri R, Battelino T, et al. International factors for mortality of adult inpatients with consensus on use of continuous glucose monitor- COVID-19 in Wuhan, China: a retrospective cohort ing. Diabetes Care. 2017;40(12):1631–40. study. Lancet. 2020;395:1054–62. 13. Battelino T, Danne T, Bergenstal RM, et al. Clinical 2. Bode B, Garrett V, Messler J, et al. Glycemic char- targets for continuous glucose monitoring data acteristics and clinical outcomes of COVID-19 interpretation: recommendations from the inter- patients hospitalized in the United States. J Dia- national consensus on time in range. Diabetes Care. betes Sci Technol. 2020;14:813–21. 2019;42(8):1593–603. 3. Chen Y, Yang D, Cheng B, et al. Clinical charac- 14. Barchetta I, Cimini FA, Bertoccini L, et al. Effects of teristics and outcomes of patients with diabetes and work status changes and perceived stress on gly- COVID-19 in association with glucose-lowering caemic control in individuals with type 1 diabetes medication. Diabetes Care. 2020;43:1399–407. during COVID-19 lockdown in Italy. Diabetes Res Clin Pract. 2020;170:108513. 4. Holman N, Knighton P, Kar P, et al. Risk factors for COVID-19-related mortality in people with type 1 15. Mesa A, Viñals C, Pueyo I, et al. The impact of strict and type 2 diabetes in England: a population-based COVID-19 lockdown in Spain on glycemic profiles cohort study. Lancet Diabetes Endocrinol. 2020;8: in patients with type 1 diabetes prone to hypo- 823–33. glycemia using standalone continuous glucose monitoring. Diabetes Res Clin Pract. 2020;167: 108354. Diabetes Ther 16. Viñals C, Mesa A, Roca D, et al. Management of and use of telemedicine on glycemic control in glucose profile throughout strict COVID-19 lock- people with type 1 diabetes during the COVID-19 down by patients with type 1 diabetes prone to outbreak in Saudi Arabia. Diabetes Res Clin Pract. hypoglycaemia using sensor-augmented pump. 2021;173:108682. Acta Diabetol. 2021;58(3):383–8. https://doi.org/ 10.1007/s00592-020-01625-2. 27. Park S-D, Kim S-W, Moon JS, et al. Impact of social distancing due to coronavirus disease 2019 on the 17. Bonora BM, Boscari F, Avogaro A, Bruttomesso D, changes in glycosylated hemoglobin level in people Fadini GP. Glycaemic control among people with with type 2 diabetes mellitus. Korean Diabetes J. type 1 diabetes during lockdown for the SARS-CoV- 2020;45:109–14. 2 outbreak in Italy. Diabetes Ther. 2020;11: 1369–79. 28. Karatas S, Yesim T, Beysel S. Impact of lockdown COVID-19 on metabolic control in type 2 diabetes 18. Prabhu Navis J, Leelarathna L, Mubita W, et al. mellitus and healthy people. Prim Care Diabetes. Impact of COVID-19 lockdown on flash and real- 2021;15:424–7. time glucose sensor users with type 1 diabetes in England. Acta Diabetol. 2020;58(2):231–7. 29. Falcetta P, Aragona M, Ciccarone A, et al. Impact of COVID-19 lockdown on glucose control of elderly 19 Capaldo B, Annuzzi G, Creanza A, et al. Blood glu- people with type 2 diabetes in Italy. Diabetes Res cose control during lockdown for COVID-19: CGM Clin Pract. 2021;174: 108750. https://doi.org/10. metrics in Italian adults with type 1 diabetes. Dia- 1016/j.diabres.2021.108750. betes Care. 2020;43:e88–9. https://doi.org/10.2337/ dc20-1127. 30. Rastogi A, Hiteshi P, Bhansali A. Improved glycemic control amongst people with long-standing dia- 20. Potier L, Hansel B, Larger E, et al. Stay-at-home betes during COVID-19 lockdown: a prospective, orders during the COVID-19 pandemic, an oppor- observational, nested cohort study. Int J Diabetes tunity to improve glucose control through behav- Dev Ctries. 2020;40:476–81. ioral changes in type 1 diabetes. Diabetes Care. 2020;44:839–43. 31. Ruissen MM, Regeer H, Landstra CP, et al. Increased stress, weight gain and less exercise in relation to 21. Schiaffini R, Barbetti F, Rapini N, et al. School and glycemic control in people with type 1 and type 2 pre-school children with type 1 diabetes during diabetes during the COVID-19 pandemic. BMJ Covid-19 quarantine: the synergic effect of parental Open Diabetes Res Care. 2021;9:e002035. care and technology. Diabetes Res Clin Pract. 2020;166:108302. 32. Verma A, Rajput R, Verma S, Balania VKB, Jangra B. Impact of lockdown in COVID 19 on glycemic 22. Dover AR, Ritchie SA, McKnight JA, et al. Assess- control in patients with type 1 diabetes mellitus. ment of the effect of the COVID-19 lockdown on Diabetes Metab Syndrome Clin Res Rev. 2020;14: glycaemic control in people with type 1 diabetes 1213–6. using flash glucose monitoring. Diabetic Med. 2021;38(1):e14374. 33. Phillip M, Bergenstal RM, Close KL, et al. The digi- tal/virtual diabetes clinic: the future is now—rec- 23. Braune K, Boss K, Schmidt-Herzel J, et al. Shaping ommendations from an international panel on workflows in digital and remote diabetes care dur- diabetes digital technologies introduction. Diabetes ing the COVID-19 pandemic via a service design: Technol Ther. 2021;23:146–54. prospective, longitudinal, open-label feasibility trial. JMIR Mhealth Uhealth. 2021;9(4):e24374. 34. Bain SC, Czernichow S, Bøgelund M, et al. Costs of COVID-19 pandemic associated with diabetes in 24. Boscari F, Ferretto S, Uliana A, Avogaro A, Brut- Europe: a health care cost model. Curr Med Res tomesso D. Efficacy of telemedicine for persons Opin. 2020;37:1–16. with type 1 diabetes during Covid19 lockdown. Nutr Diabetes. 2021;11:1. 35. Stokes EK, Zambrano LD, Anderson KN, et al. Coronavirus disease 2019 case surveillance—United 25. Rachmiel M, Lebenthal Y, Mazor-Aronovitch K, States, January 22–May 30, 2020. Morbidity Mortal et al. Glycaemic control in the paediatric and Wkly Rep. 2020;69:759–65. young adult population with type 1 diabetes fol- lowing a single telehealth visit—what have we 36. Merzon E, Green I, Shpigelman M, et al. Hae- learned from the COVID-19 lockdown? Acta Dia- moglobin A1c is a predictor of COVID-19 severity betol. 2021;58(6):697–705. in patients with diabetes. Diabetes Metabolism Res Rev. 2021;37(5):e3398. 26. Alharthi SK, Alyusuf EY, Alguwaihes AM, Alfadda A, Al-Sofiani ME. The impact of a prolonged lockdown Diabetes Ther 37. Williamson EJ, Walker AJ, Bhaskaran K, et al. Fac- after SARS-CoV-2 infection: a case report. Nat tors associated with COVID-19-related death using Metabolism. 2020;2:1021–4. OpenSAFELY. Nature. 2020;584:430–6. 49 Unsworth R, Wallace S, Oliver NS, et al. New-onset 38. Carrasco-Sánchez FJ, López-Carmona MD, Martı́- type 1 diabetes in children during COVID-19: mul- nez-Marcos FJ, et al. Admission hyperglycaemia as a ticenter regional findings in the U.K. Diabetes Care. predictor of mortality in patients hospitalized with 2020;43(11):e170–1. COVID-19 regardless of diabetes status: data from the Spanish SEMI-COVID-19 Registry. Ann Med. 50. Li H, Tian S, Chen T, et al. Newly diagnosed dia- 2020;53:103–16. betes is associated with a higher risk of mortality than known diabetes in hospitalized patients with 39. Lampasona V, Secchi M, Scavini M, et al. Antibody COVID-19. Diabetes Obes Metab. 2020;22: response to multiple antigens of SARS-CoV-2 in 1897–906. patients with diabetes: an observational cohort study. Diabetologia. 2020;63:2548–58. 51. Fadini GP, Morieri ML, Boscari F, et al. Newly-di- agnosed diabetes and admission hyperglycemia 40. Coppelli A, Giannarelli R, Aragona M, et al. predict COVID-19 severity by aggravating respira- Hyperglycemia at hospital admission is associated tory deterioration. Diabetes Res Clin Pract. with severity of the prognosis in patients hospital- 2020;168:108374. ized for COVID-19: the Pisa COVID-19 Study. Dia- betes Care. 2020;43:2345–8. 52. Garg SK, Rodbard D, Hirsch IB, Forlenza GP. Managing new-onset type 1 diabetes during the 41. Müller JA, Groß R, Conzelmann C, et al. SARS-CoV- COVID-19 pandemic: challenges and opportuni- 2 infects and replicates in cells of the human ties. Diabetes Technol Ther. 2020;22:431–9. endocrine and exocrine pancreas. Nat Metab. 2021;3(2):149–65. 53. Association of British Clinical Diabetologists. Donation of Freestyle Libre Sensors. 2020. https:// 42 Atkinson MA, Powers AC. Distinguishing the real abcd.care/announcement/donation-freestyle-libre- from the hyperglycemia: does COVID-19 induce sensors. Accessed 14 July 2021. diabetes? Lancet Diab Endocrinol. 2021. https://doi. org/10.1016/S2213-8587(21)00087-5. 54. Joensen LE, Madsen KP, Holm L, et al. Diabetes and COVID-19: psychosocial consequences of the 43. Brufsky A. Hyperglycemia, hydroxychloroquine, COVID-19 pandemic in people with diabetes in and the COVID-19 pandemic. J Med Virol. 2020;92: Denmark—what characterizes people with high 770–5. levels of COVID-19-related worries? Diabetic Med. 2020;37:1146–54. 44. Shen Y, Fan X, Zhang L, et al. Thresholds of glyce- mia and the outcomes of COVID-19 complicated 55. Solomon MD, McNulty EJ, Rana JS, et al. The Covid- with diabetes: a retrospective exploratory study 19 pandemic and the incidence of acute myocardial using continuous glucose monitoring. Diabetes infarction. N Engl J Med. 2020;383:691–3. Care. 2021;44(4):976–82. https://doi.org/10.2337/ dc20-1448. 56. Pop R, Quenardelle V, Hasiu A, et al. Impact of the COVID-19 outbreak on acute stroke pathways—in- 45. Klonoff DC, Messler JC, Umpierrez GE, et al. Asso- sights from the Alsace region in France. Eur J Neu- ciation between achieving inpatient glycemic con- rol. 2020;27:1783–7. trol and clinical outcomes in hospitalized patients with COVID-19: a multicenter, retrospective hos- 57. Lazzerini M, Barbi E, Apicella A, Marchetti F, Car- pital-based analysis. Diabetes Care. 2020;44: dinale F, Trobia G. Delayed access or provision of 578–85. care in Italy resulting from fear of COVID-19. Lan- cet Child Adolesc Health. 2020;4:e10–1. 46. Papadopoulos VP, Koutroulos M-V, Zikoudi D-G, et al. Diabetes-related acute metabolic emergencies 58. Lynn RM, Avis JL, Lenton S, Amin-Chowdhury Z, in COVID-19 patients: a systematic review and Ladhani SN. Delayed access to care and late pre- meta-analysis. Diabetol Int. 2021;23:1–15. https:// sentations in children during the COVID-19 pan- doi.org/10.1007/s13340-021-00502-9. demic: a snapshot survey of 4075 paediatricians in the UK and Ireland. Arch Dis Child. 2021;106: 47. Yang J-K, Lin S-S, Ji X-J, Guo L-M. Binding of SARS e8–e8. coronavirus to its receptor damages islets and cau- ses acute diabetes. Acta Diabetol. 2010;47:193–9. _ 59. Dzygało K, Nowaczyk J, Szwilling A, Kowalska A. Increased frequency of severe diabetic ketoacidosis 48. Hollstein T, Schulte DM, Schulz J, et al. Autoanti- at type 1 diabetes onset among children during body-negative insulin-dependent diabetes mellitus COVID-19 pandemic lockdown: an observational Diabetes Ther cohort study. Pediatric Endocrinol Diabetes Metab. 2021;38(3): e14462. https://doi.org/10.1111/dme. 2020;26:167–75. 14462. 60. Lawrence C, Seckold R, Smart C, et al. Increased 70. March CA, Flint A, DeArment D, Gilliland A, Kelly paediatric presentations of severe diabetic ketoaci- K, Rizzitano E, et al. Paediatric diabetes care during dosis in an Australian tertiary centre during the the COVID-19 pandemic: lessons learned in scaling COVID-19 pandemic. Diabetic Med. 2021;38: up telemedicine services. Endocrinol Diabetes e14417. Metab. 2021;4:e00202. 61. Ho J, Rosolowsky E, Pacaud D, et al. Diabetic 71. DeSalvo DJ, Miller KM, Hermann JM, Maahs DM, ketoacidosis at type 1 diabetes diagnosis in children Hofer SE, Clements MA, et al. Continuous glucose during the COVID-19 pandemic. Pediatr Diabetes. monitoring and glycemic control among youth 2021;22:552–7. with type 1 diabetes: International comparison from the T1D Exchange and DPV Initiative. Pediatr 62. Bornstein SR, Rubino F, Khunti K, et al. Practical Diabetes. 2018;19:1271–5. recommendations for the management of diabetes in patients with COVID-19. Lancet Diabetes Endo- 72. Gerhardsson P, Schwandt A, Witsch M, Kordonouri crinol. 2020;8:546–50. O, Svensson J, Forsander G, Battelino T, Veeze H, Danne T. The SWEET project 10-year benchmarking 63. Laffel LM, Limbert C, Phelan H, Virmani A, Wood J, in 19 countries worldwide is associated with Hofer SE. ISPAD Clinical Practice Consensus improved HbA1c and increased use of diabetes Guidelines 2018: sick day management in children technology in youth with type 1 diabetes. Diabetes and adolescents with diabetes. Pediatr Diabetes. Technol Ther. 2021. https://doi.org/10.1089/dia. 2018;19:193–204. 2020.0618. 64. Croft A, Bucca A, Jansen JH, et al. First-time diabetic 73. Cardona-Hernandez R, Schwandt A, Alkandari H, ketoacidosis in type 2 diabetics with Covid-19 et al. Glycemic outcome associated with insulin infection: a novel case series. J Emerg Med. 2020;59: pump and glucose sensor use in children and ado- e193-197. lescents with type 1 diabetes. Data from the inter- national pediatric registry SWEET. Diabetes Care. 65. Charleer S, Block CD, Huffel LV, et al. Quality of life 2021;44(5):1176–84. and glucose control after 1 year of nationwide reimbursement of intermittently scanned continu- 74. Ohannessian R, Duong TA, Odone A. Global Tele- ous glucose monitoring in adults living with type 1 medicine Implementation and Integration Within diabetes (FUTURE): a prospective observational Health Systems to Fight the COVID-19 Pandemic: A real-world cohort study. Diabetes Care. 2020;43: Call to Action. JMIR Public Heal Surveillance. 389–97. 2020;6:e18810. 66. Roussel R, Riveline J-P, Vicaut E, et al. Important 75. Demeke HB, Merali S, Marks S, Pao LZ, Romero L, drop in rate of acute diabetes complications in Sandhu P, et al. Trends in use of telehealth among people with type 1 or type 2 diabetes after initiation health centers during the COVID-19 Pandemic— of flash glucose monitoring in France: The RELIEF United States, June 26–November 6, 2020. Mor- Study. Diabetes Care. 2021. https://doi.org/10. bidity Mortal Wkly Rep. 2021;70:240–4. 2337/dc20-1690. 76. Centers for Medicare & Medicaid Services. Tele- 67. Bergenstal RM, Kerr MSD, Roberts GJ, Souto D, medicine health care provider fact sheet. Baltimore, Nabutovsky Y, Hirsch IB. Flash CGM is associated MD: US Department of Health and Human Services, with reduced diabetes events and hospitalizations Centers for Medicare & Medicaid Services; 2020. . in insulin-treated type 2 diabetes. J Endocr Soc. 2020. https://www.cms.gov/newsroom/fact-sheets/ 2021;5:bvab013. medicare-telemedicine-health-care-provider-fact- sheet. Accessed 14 July 2021. 68. Scott SN, Fontana FY, Züger T, Laimer M, Stettler C. Use and perception of telemedicine in people with 77. Christoforidis A, Kavoura E, Nemtsa A, Pappa K, type 1 diabetes during the COVID-19 pandemic— Dimitriadou M. Coronavirus lockdown effect on results of a global survey. Endocrinol Diabetes type 1 diabetes management on children wearing Metabolism. 2021;4:e00180. insulin pump equipped with continuous glucose monitoring system. Diabetes Res Clin Pract. 69. Choudhary P, Wilmot EG, Owen K, Patel DC, Mills 2020;166:108307. L, et al. A roadmap to recovery: ABCD recommen- dations on risk stratification of adult patients with 78. Fernández E, Cortazar A, Bellido V. Impact of covid- diabetes in the post-COVID-19 era. Diabet Med. 19 lockdown on glycemic control in patients with Diabetes Ther type 1 diabetes. Diabetes Res Clin Pract. 2020;166: type 1 diabetes under COVID-19 lockdown. Acta 108348. Diabetol. 2020;57:1511–7. 79. Cotovad-Bellas L, Tejera-Pérez C, Prieto-Tenreiro A, 84 Pla B, Arranz A, Knott C, et al. Impact of COVID-19 Sánchez-Bao A, Bellido-Guerrero D. The challenge lockdown on glycemic control in adults with type 1 of diabetes home control in COVID-19 times: proof diabetes mellitus. J Endocr Soc. 2020;4:bvaa149. is in the pudding. Diabetes Res Clin Pract. 2020;168:108379. 85. Predieri B, Leo F, Candia F, et al. Glycemic control improvement in Italian children and adolescents 80. Caruso I, Molfetta SD, Guarini F, et al. Reduction of with type 1 diabetes followed through telemedicine hypoglycaemia, lifestyle modifications and psy- during lockdown due to the COVID-19 pandemic. chological distress during lockdown following Front Endocrinol. 2020;11:595735. SARS-CoV-2 outbreak in type 1 diabetes. Diabetes Metab Res Rev. 2020. https://doi.org/10.1002/dmrr. 86. Sanchez-Conejero MS, Amigo JG de B, Bravo MLT, 3404 . Jiménez JM de N. Repercusión del confinamiento por COVID-19 sobre el control glucémico en niños 81. Aragona M, Rodia C, Bertolotto A, et al. Type 1 y adolescentes con diabetes mellitus tipo 1. An diabetes and COVID-19: the ‘‘Lockdown effect.’’ Pediatr (Engl Ed). 2021. https://doi.org/10.1016/j. Diabetes Res Clin Pract. 2020;108468. https://doi. anpedi.2020.12.021. org/10.1016/j.diabres.2020.108468. 87. Marigliano M, Maffeis C. Glycemic control of chil- 82. Di Dalmazi G, Maltoni G, Bongiorno C, et al. dren and adolescents with type 1 diabetes improved Comparison of the effects of lockdown due to after COVID-19 lockdown in Italy. Acta Diabetol. COVID-19 on glucose patterns among children, 2021;58(5):661–4. adolescents, and adults with type 1 diabetes: CGM study. BMJ Open Diabetes Res Care. 2020;8: 88. van der Linden J, Welsh JB, Hirsch IB, Garg SK. Real- e001664. time continuous glucose monitoring during the coronavirus disease 2019 pandemic and its impact 83. Brener A, Mazor-Aronovitch K, Rachmiel M, et al. on time in range. Diabetes Technol Ther. 2021;23:S- Lessons learned from the continuous glucose 1-S-7. monitoring metrics in pediatric patients with