Changes in metrics of continuous glucose monitoring during COVID-19 in Korean children and adolescents with type 1 diabetes mellitus

Article information

Ann Pediatr Endocrinol Metab. 2025;30(1):38-44

Publication date (electronic) : 2025 February 28

doi : https://doi.org/10.6065/apem.2448036.018

Hwa Young Kim ¹^,²

, So Hyun Shin ¹

, Hyunju Lee ¹^,²

, Jaehyun Kim^,¹^,²

¹Department of Pediatrics, Seoul National University Bundang Hospital, Seongnam, Korea

²Department of Pediatrics, Seoul National University College of Medicine, Seoul, Korea

Address for correspondence: Jaehyun Kim Department of Pediatrics, Seoul National University Bundang Hospital, 82, Gumi-ro 173 Beon-gil, Bundang-gu, Seongnam 13620, Korea Email: pedendo@snubh.org

Received 2024 February 8; Revised 2024 April 21; Accepted 2024 May 13.

Abstract

Purpose

There are limited data regarding changes in glucose control in pediatric patients with type 1 diabetes (T1D) affected by coronavirus disease 2019 (COVID-19). This study aimed to evaluate changes in the metrics of a continuous glucose monitoring (CGM) system during COVID-19 infection in children and adolescents with T1D.

Methods

Eighteen patients with T1D (<18 years of age) were included in this retrospective study. The effects of COVID-19 on CGM metrics were assessed at 5 time points (2 weeks before COVID-19 [time 1], 1 week before COVID-19 [time 2], during COVID-19 [time 3], 1 week after COVID-19 [time 4], and 2 weeks after COVID-19 [time 5]).

Results

All participants had at least 1 symptom of COVID-19 and did not need to be hospitalized. The glucose management indicator (GMI) was higher at time 3 (7.7%±1.4%) compared to time 1 (7.1%±1.1%; P=0.016) and time 5 (7.0%±1.2%; P=0.008). According to the insulin delivery method, the GMI at time 3 was significantly higher than that at time 5 in patients treated with multiple daily injections (MDI) (median and interquartile range, 8.0% [6.1%–8.5%] vs. 7.1% [5.8%–7.9%]; P=0.020) but not in those treated with continuous subcutaneous insulin infusion (CSII).

Conclusions

Pediatric patients with T1D and mild COVID-19 showed worsening glycemic control during COVID-19 infection, but it returned to preinfection levels within 2 weeks of infection. CSII is more effective in maintaining stable glycemic control during COVID-19 infection than is MDI therapy.

Keywords: COVID-19; Continuous glucose monitoring; Adolescent; Child; Type 1 diabetes mellitus; Korea

Highlights

· Pediatric type 1 diabetes patients experienced worsened glycemic control during coronavirus disease 2019, with a significant increase in glucose management indicator levels. However, glucose levels returned to preinfection values within 2 weeks. Continuous subcutaneous insulin infusion was more effective than multiple daily injections in maintaining stable glucose control.

Introduction

Considering the initial outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in December 2019 in China, the World Health Organization declared coronavirus disease 2019 (COVID-19) a global pandemic on March 11, 2020 [1]. Nearly 15.6 million children were reported to have tested positive for COVID-19 by May 11, 2023, representing 17.9% of all pediatric cases [2]. Children affected by COVID-19 present generally milder symptoms and have better outcomes than adults [3,4]. However, those with comorbidities such as diabetes mellitus appear to be at a higher risk of severe illness [5,6]. Because the pediatric age group was less commonly affected by COVID-19 than the adult population, information is lacking on the clinical course in community-infected pediatric patients with type 1 diabetes (T1D).

The COVID-19 pandemic lockdowns led to changes in lifestyle and routine care for T1D [7,8]. The use of technologies such as continuous glucose monitoring (CGM) was recommended for children and adolescents with T1D to maintain good metabolic control during the COVID-19 pandemic [9]. CGM use is more effective than self-monitoring of blood glucose in observing glucose trends, making it particularly valuable for daily care [10]. During the pandemic, the prevalence and severity of diabetic ketoacidosis (DKA) increased in children with new-onset T1D [11-13]. In contrast, prior studies in the United States and Europe indicate that children and adolescents with pre-existing T1D who used CGM showed improved CGM metrics during the lockdown period compared to the pre-pandemic period [14-18].

Illnesses can increase blood glucose levels in patients with pre-existing T1D by increasing stress hormone levels, which promote gluconeogenesis and glycogenolysis [19]. An observational study of 36 CGM users at 3 pediatric diabetes clinics in Israel reported increased time in hyperglycemia and a trend of reduced time-in-target glycemic range during a short infection period [20]. The changes in glucose metrics varied based on the insulin treatment regimen in 32 Italian young adults with T1D affected by COVID-19 [21]. Currently, data regarding the effects of COVID-19 on CGM metrics during active COVID-19 infection in pediatric patients with T1D are limited. In addition, it is not known whether CGM metrics differ according to insulin delivery method in pediatric patients with T1D. This study aimed to evaluate the changes in CGM metrics and differences according to insulin delivery method during COVID-19 infection in Korean children and adolescents with T1D.

Materials and methods

1. Study participants

Thirty-three children and adolescents under 18 years of age with T1D who were infected with COVID-19 and visited Seoul National University Bundang Hospital between February 2022 and November 2022 were included. Among them, 15 were excluded because of unavailable CGM data (n=12), missing questionnaires (n=1), or use of corticosteroids during infection (n=2). The remaining 18 patients with T1D using CGM with sensor use ≥70% of the time for at least 5 weeks during COVID-19 infection were included in the analysis (Fig. 1). The diagnosis of T1D was defined by the presence of one or more islet autoantibodies and/or a decreased plasma C-peptide level [22,23] All but 2 patients were treated with intensive insulin therapy: 2 (11.1%) conventional, 10 (55.6%) MDI, and 6 (33.3%) CSII regimens. All study participants were using rtCGM (G6, Dexcom Inc., San Diego, CA, USA) or isCGM (FreeStyle, Abbott Diabetes Care, Abbott Park, IL, USA). Fifteen patients (83.3%) were using real-time CGM (rtCGM), and 3 (16.7%) were using intermittently scanned CGM (isCGM).

Fig. 1.

Flowchart of the study population. T1D, type 1 diabetes; COVID-19, coronavirus disease 2019; CGM, continuous glucose monitoring.

2. Measurements

Medical records regarding demographics, glycated hemoglobin (HbA1c) levels, and the presence of micro- or macrovascular complications were retrospectively reviewed. Information about COVID-19 (time of diagnosis, symptoms during infection, and related therapy) was collected using a questionnaire as part of routine clinical practice for COVID-19-infected patients. CGM data were collected from data uploaded on the following web-based platforms: Clarity (https://clarity.dexcom.eu/professional/) and Libreview (https://libreview.com/chooseCountryLanguage). Data were retrieved for 5 1-week periods based on the timing of symptom onset: time 1 (2 weeks before COVID-19), time 2 (1 week before COVID-19), time 3 (during COVID-19), time 4 (1 week after COVID-19), and time 5 (2 weeks after COVID-19). For each period, the following CGM-related metrics were collected: average sensor glucose, glucose management indicator (GMI), standard deviation (SD), coefficient of variation, percentage of time in the target range (TIR, 70–180 mg/dL), percentages of time above range (TAR) level 1 between 180 and 250 mg/dL and above TAR level 2 >250 mg/dL, and percentages of time spent below range (TBR) level 1 between 54 and 69 mg/dL, and below TBR level 2 <54 mg/dL [24].

3. Statistical analysis

All statistical analyses were performed using IBM SPSS Statistics ver. 27.0 (IBM Co., Armonk, NY, USA). The continuous variables are presented as mean±SD for variables that follow a normal distribution, and median (interquartile range [IQR]) for skewed variables. The categorical variables are shown as counts and percentages. Repeated-measure analysis of variance was performed for each CGM variable with Tukey test correction, if necessary. The Friedman test was used as a nonparametric method to analyze changes in glucose metrics according to insulin delivery method. Statistical significance was set at a P-value < 0.05.

4. Ethical statement

The Institutional Review Board of Seoul National University Bundang Hospital approved the study protocol and waived the requirement for informed consent (approval number: B-2302-811-105).

Results

1. Participant characteristics

Table 1 shows the baseline characteristics of 18 patients with T1D (6 males, 12 females). The mean age at T1D diagnosis was 7.2±4.0 years. None of the patients had any micro- or macrovascular complications. The mean age at the time of COVID-19 infection was 10.2±3.9 years, and the mean disease duration was 4.1±3.4 years. The mean HbA1c level prior to COVID-19 infection was 7.0%±1.4%. Six patients (33.3%) had been vaccinated against COVID-19. All patients had at least 1 COVID-19-related symptom: fever in 18 (100%), sore throat in 13 (72.2%), cough in 9 (50.0%), fatigue in 8 (44.4%), rhinorrhea in 7 (38.9%), headache in 6 (33.3%), myalgia in 5 (27.8%), loss of smell or taste in 2 (11.1%), abdominal pain in 1 (5.6%), and diarrhea in 1 (5.6%). The mean duration of symptoms of COVID-19 was 5.3±3.9 days. No patient required hospitalization.

Table 1.

Clinical characteristics of patients

2. Changes in glucose metrics over time

The average sensor glucose level at time 3 (178.8±45.9 mg/dL) was significantly higher than those at time 1 (159.0±36.7 mg/dL, P=0.010) and time 5 (157.4±38.1 mg/dL, P=0.004). Likewise, the GMI at time 3 (7.7%±1.4%) was higher than were those at time 1 (7.1%±1.1%, P=0.016) and time 5 (7.0%±1.2%, P=0.008) (Table 2). The change in TAR1 for each of the 5 time points was significant (P=0.046), but the differences between groups were not significant (Fig. 2). No significant changes were found in the measured HbA1c levels before and after COVID-19 infection (7.0%±1.4% and 6.9%±0.9%, respectively).

Table 2.

Changes in glycemic metrics in different time points

Fig. 2.

Changes in glycemic metrics over time. TBR2, time below range level 2; TBR1, time below range level 1; TIR, time in range; TAR1, time above range level 1; TAR2, time above range level 2.

3. Changes in glucose metrics according to insulin delivery method

Table 3 shows changes in glucose metrics according to insulin delivery method. The average sensor glucose levels (median and IQR, 183.0 mg/dL [127.8–203.8 mg/dL] vs. 156.0 mg/dL [119.8–178.3 mg/dL]; P=0.018) and GMI (median and IQR, 8.0% [6.1%–8.5%] vs. 7.1% [5.8%–7.9%]; P=0.020) were significantly higher at time 3 compared to time 5 in patients treated with MDI (n=10), but this difference was not observed in those using CSII (n=6). Participants with CSII showed no significant changes in glucose metrics before and after COVID-19 infection.

Table 3.

Changes in glucose metrics according to insulin delivery methods

Discussion

In the present study, pediatric patients with T1D had a mild course of COVID-19. Worsening of the average sensor glucose and GMI levels were observed during COVID-19 infection, but both returned to pre-infection levels within 2 weeks after infection. The average sensor glucose levels and GMI were higher during COVID-19 infection compared to 2 weeks after COVID-19 infection in patients treated with MDI, but this difference was not observed in those using CSII.

Although the clinical manifestations of COVID-19 in children range from asymptomatic or mild to critical illness, they are generally milder than symptoms in adults; this discrepancy seems to be associated with limited exposure to the virus among children and host factors [25-29]. In accordance with previous reports on the pediatric population affected by COVID-19, our pediatric patients with pre-existing T1D also had mild symptoms, lasting an average of 5.3 days. Fever, sore throat, and cough were the most common symptoms (observed in more than 50% of patients). A large retrospective study in the United States reported that children with T1D hospitalized for COVID-19 developed more severe illnesses than did nondiabetic children [30]. However, a large cohort study that followed 17,110 children and adolescents with T1D in 4 countries (China, Italy, Spain, and United States) reported only 1 of 16 confirmed COVID-19 cases requiring admission to the intensive care unit to manage DKA, which suggests that children with T1D had similar outcomes and prognoses to their nondiabetic peers [31].

In our pediatric patients aged 2–17 years with T1D and symptomatic COVID-19 infection, significant changes in the average sensor glucose levels and GMI were observed during COVID-19 infection compared to the preinfection period, while no significant changes were found in measured HbA1c. The average sensor glucose levels and GMI were significantly higher during COVID-19 than they were 2 weeks before COVID-19; these levels returned to preinfection levels 2 weeks after COVID-19. Although limited data are available on pediatric patients with T1D and COVID-19 using CGM, a retrospective study of 36 young Israelis patients with T1D using CGM reported a similar temporary increase of average sensor glucose during the short infection period (from 161.6±27.6 mg/dL to 169.4±30.6 mg/dL) without significant changes in HbA1c levels before and after infection [20]. In addition, a temporary increase in TAR1 (from 32.2% to 36.1%) but decreases in TBR1 and TBR2 (from 3.0% to 1.4% and from 0.6% to 0.0%, respectively) were observed during COVID-19 infection compared to the preinfection period in that study [20]. Although the factors that worsen glucose metrics during COVID-19 are not well known, the observed increase in sensor glucose levels could be explained by the stress response triggered by the COVID-19 viral infection. However, the absence of significant changes in HbA1c before and after infection suggests that COVID-19 infection does not have a long-term effect on glucose metabolism.

In our analysis, according to the insulin delivery method, higher average sensor glucose levels and GMI were observed during COVID 19 compared to those 2 weeks after COVID-19 in the MDI group but not in the CSII group. A study of Italian young adults with T1D also reported more pronounced deterioration in glucose control and variability during COVID-19 in patients treated with MDI than in those treated with CSII [21]. These findings are in line with those of previous studies that reported that CSII is more effective in lowering glucose variability and increasing TIR than is MDI therapy in patients with T1D [32,33]. Advancements in diabetes-related technologies, such as CGM and automated insulin delivery devices, have played an important role in the management of T1D. These technologies have become more emphasized during the COVID-19 pandemic. During COVID-19, the risk of DKA may be lowered by adjusting insulin dose using the trend arrows on the CGM device or changing the mode of automated insulin delivery devices in addition to following the sick day management guidelines [9,10].

This study has several limitations. First, a relatively small number of patients was included. In addition, information on the total daily insulin dose, which could affect changes in glucose metrics during the infection period, could not be evaluated owing to the retrospective design of the study. Second, the selection of patients from a single tertiary center and the self-reported information on COVID-19 diagnosis and clinical symptoms have the potential for bias. In addition, as our study was limited to patients with symptomatic cases using CGM who attended scheduled clinic visits, the results are not representative of all children and adolescents with T1D. Nevertheless, to the best of our knowledge, this study is the first to evaluate the effect of COVID-19 on glucose metrics in children and adolescents with pre-existing T1D. Our study was strengthened by the comparison of real-world serial CGM data before, during, and after COVID-19 in a pediatric population with T1D.

In conclusion, Korean children and adolescents with T1D experienced mild COVID-19 symptoms similar to those of the general pediatric population. Worsening of glucose metrics was observed during COVID-19, which returned to preinfection levels within 2 weeks after infection. The use of CSII may provide additional benefits for maintaining stable glycemic control during COVID-19. Further studies on the factors positively affecting glucose metrics during COVID-19 in pediatric patients with T1D may be helpful in optimizing glycemic management.

Notes

Conflicts of interest

No potential conflict of interest relevant to this article was reported.

Funding

This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sector.

Data availability

The data that support the findings of this study can be provided by the corresponding author upon reasonable request.

Author Contribution

Conceptualization: HYK, HL, JK; Data curation: HYK, SHS, HL, JK; Formal analysis: HYK, SHS, JK; Methodology: HYK, HL, JK; Visualization: HYK, SHS, JK; Writing – original draft: HYK; Writing – review & editing: HYK, SHS, HL, JK

References

1. World Health Organization. WHO COVID-19 dashboard. Number of COVID-19 cases reported to WHO [Internet]. Geneva (Switzerland): WHO; 2025. [updated 2024 Jan 21; cited 2023 Oct 9]. Available from: https://data.who.int/dashboards/covid19/cases?n=c.

2. American Academy of Pediatrics. Children and COVID-19: state level data report [Internet]. Itasca (IL): American Academy of Pediatrics; [updated 2023 May 11; cited 2024 Apr 9]. Available from: https://downloads.aap.org/AAP/PDF/AAP%20COVID-19%20State%20Data%20Report%205.11.23%20FINAL.pdf.

3. Jiehao C, Jin X, Daojiong L, Zhi Y, Lei X, Zhenghai Q, et al. A case series of children with 2019 novel coronavirus infection: clinical and epidemiological features. Clin Infect Dis 2020;71:1547–51.

4. Cavallo F, Rossi N, Chiarelli F. Novel coronavirus infection and children. Acta Biomed 2020;91:172–6.

5. Shen KL, Yang YH, Jiang RM, Wang TY, Zhao DC, Jiang Y, et al. Updated diagnosis, treatment and prevention of COVID-19 in children: experts' consensus statement (condensed version of the second edition). World J Pediatr 2020;16:232–9.

6. Choi JH, Choi SH, Yun KW. Risk factors for severe COVID-19 in children: a systematic review and meta-analysis. J Korean Med Sci 2022;37e35.

7. Kaushal T, Tinsley LJ, Volkening LK, Turcotte C, Laffel LM. Improved CGM glucometrics and more visits for pediatric type 1 diabetes using telemedicine during 1 year of COVID-19. J Clin Endocrinol Metab 2022;107:e4197–202.

8. Kang S, Seo MY, Kim SH, Park MJ. Changes in lifestyle and obesity during the COVID-19 pandemic in Korean adolescents: based on the Korea Youth Risk Behavior Survey 2019 and 2020. Ann Pediatr Endocrinol Metab 2022;27:281–8.

9. d'Annunzio G, Maffeis C, Cherubini V, Rabbone I, Scaramuzza A, Schiaffini R, et al. Caring for children and adolescents with type 1 diabetes mellitus: Italian Society for Pediatric Endocrinology and Diabetology (ISPED) statements during COVID-19 pandemia. Diabetes Res Clin Pract 2020;168:108372.

10. Phelan H, Hanas R, Hofer SE, James S, Landry A, Lee W, et al. Sick day management in children and adolescents with diabetes. Pediatr Diabetes 2022;23:912–25.

11. Prosperi S, Chiarelli F. COVID-19 and diabetes in children. Ann Pediatr Endocrinol Metab 2022;27:157–68.

12. Lee Y, Kim M, Oh K, Kang E, Rhie YJ, Lee J, et al. Comparison of initial presentation of pediatric diabetes before and during the coronavirus disease 2019 pandemic era. J Korean Med Sci 2022;37e176.

13. Lee DH, Kim HY, Park JY, Kim J, Park JH. New-onset type 1 and type 2 diabetes among Korean youths during the COVID-19 pandemic. JAMA Pediatr 2025;179:155–62.

14. Christoforidis A, Kavoura E, Nemtsa A, Pappa K, Dimitriadou M. Coronavirus lockdown effect on type 1 diabetes management omicronn children wearing insulin pump equipped with continuous glucose monitoring system. Diabetes Res Clin Pract 2020;166:108307.

15. Abdulhussein FS, Chesser H, Boscardin WJ, Gitelman SE, Wong JC. Youth with type 1 diabetes had improvement in continuous glucose monitoring metrics during the COVID-19 pandemic. Diabetes Technol Ther 2021;23:684–91.

16. Di Dalmazi G, Maltoni G, Bongiorno C, Tucci L, Di Natale V, Moscatiello S, et al. Comparison of the effects of lockdown due to COVID-19 on glucose patterns among children, adolescents, and adults with type 1 diabetes: CGM study. BMJ Open Diabetes Res Care 2020;8e001664.

17. Predieri B, Leo F, Candia F, Lucaccioni L, Madeo SF, Pugliese M, et al. Glycemic control improvement in Italian children and adolescents with type 1 diabetes followed through telemedicine during lockdown due to the COVID-19 pandemic. Front Endocrinol (Lausanne) 2020;11:595735.

18. Brener A, Mazor-Aronovitch K, Rachmiel M, Levek N, Barash G, Pinhas-Hamiel O, et al. Lessons learned from the continuous glucose monitoring metrics in pediatric patients with type 1 diabetes under COVID-19 lockdown. Acta Diabetol 2020;57:1511–7.

19. Walker M, Marshall SM, Alberti KG. Clinical aspects of diabetic ketoacidosis. Diabetes Metab Rev 1989;5:651–63.

20. Nimri R, Rachmiel M, Yackobovitch-Gavan M, Brener A, de Vries L, Fisch Shvalb N, et al. Symptoms and glycemic control in young people with type 1 diabetes following SARS-CoV-2 infection: an observational study. J Clin Endocrinol Metab 2022;107:e3264–72.

21. Longo M, Scappaticcio L, Petrizzo M, Castaldo F, Sarnataro A, Forestiere D, et al. Glucose control in home-isolated adults with type 1 diabetes affected by COVID-19 using continuous glucose monitoring. J Endocrinol Invest 2022;45:445–52.

22. American Diabetes Association Professional Practice Committee. 2. Diagnosis and classification of diabetes: standards of care in diabetes-2024. Diabetes Care 2024;47:S20–42.

23. Nieto T, Castillo B, Nieto J, Redondo MJ. Demographic and diagnostic markers in new onset pediatric type 1 and type 2 diabetes: differences and overlaps. Ann Pediatr Endocrinol Metab 2022;27:121–5.

24. Battelino T, Danne T, Bergenstal RM, Amiel SA, Beck R, Biester T, et al. Clinical targets for continuous glucose monitoring data interpretation: recommendations from the international consensus on time in range. Diabetes Care 2019;42:1593–603.

25. Qiu H, Wu J, Hong L, Luo Y, Song Q, Chen D. Clinical and epidemiological features of 36 children with coronavirus disease 2019 (COVID-19) in Zhejiang, China: an observational cohort study. Lancet Infect Dis 2020;20:689–96.

26. Assaker R, Colas AE, Julien-Marsollier F, Bruneau B, Marsac L, Greff B, et al. Presenting symptoms of COVID-19 in children: a meta-analysis of published studies. Br J Anaesth 2020;125:e330–2.

27. Ludvigsson JF. Systematic review of COVID-19 in children shows milder cases and a better prognosis than adults. Acta Paediatr 2020;109:1088–95.

28. Dong Y, Mo X, Hu Y, Qi X, Jiang F, Jiang Z, et al. Epidemiology of COVID-19 among children in China. Pediatrics 2020;145e20200702.

29. Mansourian M, Ghandi Y, Habibi D, Mehrabi S. COVID-19 infection in children: a systematic review and meta-analysis of clinical features and laboratory findings. Arch Pediatr 2021;28:242–8.

30. Kompaniyets L, Agathis NT, Nelson JM, Preston LE, Ko JY, Belay B, et al. Underlying medical conditions associated with severe COVID-19 illness among children. JAMA Netw Open 2021;4e2111182.

31. Cardona-Hernandez R, Cherubini V, Iafusco D, Schiaffini R, Luo X, Maahs DM. Children and youth with diabetes are not at increased risk for hospitalization due to COVID-19. Pediatr Diabetes 2021;22:202–6.

32. Longo M, Caruso P, Petrizzo M, Castaldo F, Sarnataro A, Gicchino M, et al. Glycemic control in people with type 1 diabetes using a hybrid closed loop system and followed by telemedicine during the COVID-19 pandemic in Italy. Diabetes Res Clin Pract 2020;169:108440.

33. Keyu G, Jiaqi L, Liyin Z, Jianan Y, Li F, Zhiyi D, et al. Comparing the effectiveness of continuous subcutaneous insulin infusion with multiple daily insulin injection for patients with type 1 diabetes mellitus evaluated by retrospective continuous glucose monitoring: a real-world data analysis. Front Public Health 2022;10:990281.

Article information Continued

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Variable	Value
Male sex	6 (33.3)
Current age (yr)	10.2±3.9
Age at T1D diagnosis (yr)	7.2±4.0
Diabetes duration (yr)	4.1±3.4
HbA1c before infection (%)	7.0±1.4
Insulin delivery
Conventional	2 (11.1)
Multiple daily insulin injection	10 (55.6)
Continuous subcutaneous insulin infusion	6 (33.3)
CGM device
Real-time	15 (83.3)
Intermittently scanned	3 (16.7)

	Time 1	Time 2	Time 3	Time 4	Time 5
Average glucose (mg/dL)	159.0±36.7	167.1±34.4	178.8±45.9^*	168.1±35.9	157.4±38.1
GMI (%)	7.1±1.1	7.4±1.1	7.7±1.4^*	7.4±1.2	7.0±1.2
CV (%)	36.4±8.4	35.9±7.9	34.7±7.5	37.0±7.8	37.1±8.5
TIR (%)	65.8±19.8	60.5±21.6	58.2±23.1	62.7±19.1	66.0±20.4
TAR1 (%)	22.2±11.3	25.0±12.7	23.8±11.8	21.7±10.7	18.8±11.2
TAR2 (%)	6.8 (1.6–14.8)	7.4 (4.1–23.1)	11.3 (2.7–28.0)	9.0 (3.3–28.3)	5.9 (1.5–18.8)
TBR1 (%)	1.8 (0.35–4.15)	2.5 (0.3–4.4)	1.2 (0.3–3.3)	2.4 (0.3–5.0)	1.8 (0.7–5.6)
TBR2 (%)	0.3 (0.0–1.1)	0.1 (0.0–0.6)	0.1 (0.0–0.3)	0.1 (0.0–1.3)	0.2 (0.0–1.4)

Table 3.

Changes in glucose metrics according to insulin delivery methods

Variable	MDI group (n=10)					CSII group (n=6)
Variable	Time 1	Time 2	Time 3	Time 4	Time 5	Time 1	Time 2	Time 3	Time 4	Time 5
Average glucose (mg/dL)	156.0 (124.0–181.5)	175.5 (135.0–207.0)	183.0 (127.8–203.8)	156.0 (129.0–203.5)	156.0 (119.8–178.3)^*	154.7 (135.5–175.4)	156.5 (140.6–169.7)	165.5 (151.0–182.3)	170.6 (143.3–180.7)	146.0 (124.0–160.5)
GMI (%)	7.1 (6.0–7.9)	7.8 (6.4–8.6)	8.0 (6.1–8.5)	7.1 (6.2–8.7)	7.1 (5.8–7.9)^*	7.0 (6.4–7.7)	7.1 (6.5–7.6)	7.4 (6.9–8.0)	7.6 (6.6–7.9)	6.7 (6.0–7.2)
CV (%)	35.0 (30.0–40.7)	32.0 (28.1–40.3)	33.9 (28.8–41.3)	33.9 (30.8–40.7)	37.6 (27.0–44.2)	29.7 (27.7–45.3)	37.8 (30.1–42.2)	30.3 (26.8–43.9)	33.0 (30.7–46.3)	36.3 (26.9–41.7)
TIR (%)	65.9 (48.3–84.1)	52.1 (37.3–81.6)	52.6 (42.1–83.5)	65.6 (42.5–81.6)	61.1 (46.9–84.8)	67.3 (59.5–85.3)	70.9 (59.9–76.2)	62.0 (53.4–73.3)	61.6 (57.2–73.1)	75.4 (65.9–84.1)
TAR1 (%)	21.2 (8.4–29.5)	25.5 (11.6–34.2)	20.8 (9.2–26.9)	19.0 (10.9–25.3)	16.4 (8.4–32.4)	27.5 (14.3–34.7)	24.0 (15.4–36.0)	31.7 (24.5–39.8)	32.0 (18.0–38.3)	19.4 (9.1–26.3)
TAR2 (%)	7.9 (1.7–18.9)	15.2 (2.8–23.5)	19.9 (0.8–28.2)	4.4 (1.4–32.4)	8.1 (1.4–20.4)	6.3 (0.2–10.1)	5.9 (3.6–9.6)	7.6 (5.2–15.1)	9.8 (6.4–15.4)	4.5 (0.5–8.0)
TBR1 (%)	1.8 (0.4–2.8)	1.2 (0.0–3.3)	1.2 (0.4–2.2)	1.6 (0.6–5.4)	1.6 (0.8–5.6)	0.5 (0.2–3.4)	2.7 (0.3–3.7)	0.4 (0.1–4.3)	1.8 (0.2–4.6)	2.5 (1.4–5.4)
TBR2 (%)	0.1 (0.0–0.8)	0.1 (0.0–0.6)	0.1 (0.0–0.3)	4.4 (1.4–32.4)	0.3 (0.0–1.6)	0.3 (0.0–0.7)	0.2 (0.0–0.7)	0.0 (0.0–0.7)	9.8 (6.4–15.4)	0.2 (0.0–1.4)

Values are expressed as median (interquartile range).

MDI, multiple daily injection; CSII, continuous subcutaneous insulin infusion regimen; GMI, glucose management indicator; CV, coefficient of variation; TIR, time in range; TAR1, time above range level 1; TAR2, time above range level 2; TBR1, time below range level 1; TBR2, time below range level 2.

P<0.05 between time 3 and time 5.