Abstract
Aims/Introduction
An early-morning elevation of blood glucose levels known as the dawn phenomenon and consequent postbreakfast hyperglycemia occur in some individuals with type 1 diabetes (T1D). Whereas insulin pump therapy can mitigate this phenomenon, some individuals prefer or are limited to alternative treatments. We have now assessed the effectiveness of early-morning administration of rapid-acting insulin for amelioration of the dawn phenomenon in individuals with T1D.
Materials and Methods
Thirteen individuals with T1D who experienced the dawn phenomenon as determined by continuous glucose monitoring (CGM) and who received a small dose of rapid-acting insulin on waking were included in this retrospective study. We evaluated the change in sensor glucose levels during a 2-h period from before to after breakfast consumed at 0700 h. The change in blood glucose levels during additional time intervals, average daily sensor glucose values, CGM indices, and insulin dose were also evaluated.
Results
The early-morning administration of 0.5–1 unit of rapid-acting insulin was associated with a significant reduction in 2-h glucose variability between before (0700 h) and after breakfast from a median of 90.7–51.0 mg/dL. The glucose variability from 0300 to 0700 or 0900 h was also significantly decreased, from 67.7 to 29.0 mg/dL and from 172.5 to 78.3 mg/dL, respectively. Average sensor glucose levels throughout the day were significantly reduced (from 192.7 to 156.7 mg/dL), as was the daily total insulin dose.
Conclusion
Early-morning administration of rapid-acting insulin effectively managed the dawn phenomenon and subsequent postbreakfast hyperglycemia in individuals with T1D.
Supplementary Information
The online version contains supplementary material available at 10.1007/s13340-024-00709-6.
Keywords: Type 1 diabetes, Dawn phenomenon, Early morning, Rapid-acting insulin, Continuous glucose monitoring, Postbreakfast hyperglycemia
Introduction
The dawn phenomenon, first described by Schmidt and colleagues in 1981, is characterized by a pronounced rise in blood glucose levels during the early morning that has been attributed to suboptimal fulfillment of insulin requirements in individuals with diabetes mellitus [1]. It is distinguished from the Somogyi effect, which is a subsequent rise in blood glucose resulting from nocturnal hypoglycemia [2]. This increase in blood glucose has been found to range from 15 to ≥ 25 mg/dL, corresponding to the transition from the nadir of nocturnal glucose concentration to the peak prebreakfast level [3]. The dawn phenomenon also gives rise to increased postprandial glycemia [4, 5]. Its prevalence appears to be highest in individuals with type 1 diabetes (T1D), characterized by a pronounced deficiency in insulin secretion, and the question of how to adequately address the increased insulin demand during the early morning has presented a challenge [6, 7].
Individuals with T1D treated with a regimen of multiple daily insulin injections (MDI) experience nocturnal hypoglycemia if they attempt to satisfy the insulin requirement associated with the dawn phenomenon. Such individuals have therefore been left with no alternative but to endure the dawn phenomenon and subsequent hyperglycemia, leading to difficulties in glycemic control. On the other hand, individuals with T1D receiving insulin pump therapy can increase the insulin infusion rate at the necessary time, thus administering an adequate insulin dose to meet the early-morning demand. Insulin pump therapy has therefore long been advocated for persons with T1D who experience the dawn phenomenon [8]. However, partly due to its complexity and cost, some individuals with T1D are unable or prefer not to adopt such treatment and instead opt for MDI. Indeed, evidence suggests that only ~ 50% of individuals with T1D who receive specialized medical attention undergo insulin pump therapy [9].
The aim of the present study was to determine whether the dawn phenomenon can be effectively managed without resort to an insulin pump. Specifically, we investigated whether a minimal dose of rapid-acting insulin administered on waking might fulfill the early-morning insulin requirement and suppress the postbreakfast increase in blood glucose levels in individuals with T1D who experience the dawn phenomenon and are receiving MDI treatment.
Materials and methods
Study design and participants
This study was designed as a retrospective observational investigation to be conducted at a single center. We recruited individuals diagnosed with T1D who were admitted to the Department of Diabetes and Endocrinology at Kobe University Hospital during the period from 1 January 2010 to 30 November 2018. Such individuals who experienced the dawn phenomenon, as identified by the attending physician on the basis of subcutaneous continuous glucose monitoring (CGM) during their hospital stay, were eligible for the study. Initiation of the administration of rapid-acting insulin on awakening to address the dawn phenomenon as well as CGM for a minimum of 1 day before and 1 day after the start of such treatment were also inclusion criteria. Individuals who declined participation on the basis of published information, who were receiving insulin pump therapy, who were < 20 years of age at the time of potential enrollment, who had severe hepatic or renal dysfunction, who were undergoing treatment with steroids, or who were pregnant were excluded from the study. All procedures performed in this study of human participants were in accordance with the guidelines of the Declaration of Helsinki, and the study protocol was approved by the ethics committee of Kobe University Graduate School of Medicine (approval no. 180249).
Procedures
A well-balanced dietary regimen was provided for the therapeutic management of individuals with T1D admitted to the Department of Diabetes and Endocrinology, Kobe University Hospital. This nutritional regimen delivered 25–30 kcal per kilogram of ideal body weight and comprised 50–60% carbohydrates, 20–25% fat, and 15–20% protein. Three meals of equal caloric content were scheduled for 0700, 1200, and 1800 h. Individuals were also provided with an advanced CGM system, either Medtronic MiniMed CGMS-Gold (Medtronic, Dublin, Ireland), iPro2 (Medtronic), or FreeStyle Libre Pro (Abbott Diabetes Care, Alameda, CA, USA). Based on the CGM data, the treating endocrinologist verified fluctuations in glycemia, accordingly modulated insulin dosage, and refined the basal insulin regimen to ensure glycemic equilibrium from midnight to 0300 h, without causing hypoglycemia. Individuals with a blood glucose rise of 30 mg/dL or more from 0300 to pre-breakfast were identified as having the dawn phenomenon by the attending physician. Individuals who manifested the dawn phenomenon received either 0.5 or 1 unit of rapid-acting insulin tailored to their personalized waking schedule in the home setting was it the dose that was tailored or the time of its administration. Regarding the insulin dose, whether it be 0.5 units or 1 unit, the decision was made by the attending physician. We extracted data from the medical records of such individuals who were treated with rapid-acting insulin (insulin lispro, insulin aspart, or insulin glulisine) on awakening to target the dawn phenomenon. Glycemic variation was evaluated over a span of 1–3 days both before and after commencement of such insulin therapy, with a concurrent analysis of the overall insulin dosage administered being performed. For the analysis, data from the 1 to 3 days were converted into daily averages for each individual and then analyzed accordingly.
Outcomes
The primary endpoint of the study was to ascertain the difference in sensor glucose (SG) levels between before (0700 h) and after (0900 h) breakfast and thereby to evaluate the amplitude of glycemic fluctuations within the initial postprandial period with or without the early-morning administration of rapid-acting insulin. Secondary endpoints included assessment of the variability of blood glucose between 0300 and 0700 h, blood glucose fluctuations from 0300 to 0900 h, and the aggregated daily SG levels before and after the initiation of treatment with rapid-acting insulin on awakening. Supplementary parameters examined included CGM indices such as the standard deviation (SD), coefficient of variation, mean amplitude of glycemic excursions (MAGE), time in range, time below range, time above range, total insulin dose, and basal insulin dose.
Statistical analysis
Data showing a normal distribution were to be expressed as means ± SD, whereas those deviating from normality were to be expressed as median values accompanied by the interquartile range. Statistical analysis was performed with EZR version 1.41 software [10]. Both primary and secondary endpoints were analyzed with the Wilcoxon signed-rank test. A P value of < 0.05 was considered statistically significant.
Results
During the target period, 342 individuals with T1D were hospitalized and evaluated by CGM. Among these individuals, 152 were treated with an insulin pump and were therefore excluded from the study. The remaining 190 individuals received MDI treatment, and 16 of these individuals manifested the dawn phenomenon and received early-morning insulin administration. After exclusion of 3 individuals with insufficient data, the analyses of the present study were conducted with the remaining 13 individuals, comprising 6 men and 7 women (Fig. 1). The baseline characteristics of the study participants are presented in Table 1. The median age was 68 years (interquartile range of 58–74 years), the median disease duration was 12 years (7–21 years), and the median hemoglobin A1c level was 8.3% (7.3–8.9%). The median total insulin dose, adjusted to ensure no glucose fluctuations from midnight to 0300 h, was 30.0 units (23.2–34.0 units) on admission. Present the basal insulin dose for each individual, along with the early-morning administration of rapid-acting insulin and the timing of these administrations, as supplementary information (Supplementary Table S1).
Fig. 1.
Derivation of the study population. T1D type 1 diabetes, CGM continuous glucose monitoring, CSII continuous subcutaneous insulin infusion, MDI multiple daily insulin injections
Table 1.
Baseline characteristics of the study participants (n = 13)
| Characteristic | ||||
|---|---|---|---|---|
| Male/female | 6 | / | 7 | |
| Age (years) | 68 | (58–74) | ||
| Body mass index (kg/m2) | 22.3 | (19.8–23.9) | ||
| Hemoglobin A1c (%) | 8.3 | (7.3–8.9) | ||
| Glycated albumin (%) | 24.3 | (21.4–28.0) | ||
| Serum C-peptide (ng/mL) | 0.01 | (0.01–0.01) | ||
| Serum triglyceride (mg/dL) | 67 | (48–76) | ||
| Serum HDL-cholesterol (mg/dL) | 61 | (51–87) | ||
| Serum LDL-cholesterol (mg/dL) | 85 | (76–101) | ||
| Aspartate aminotransferase (U/L) | 21 | (17–21) | ||
| Alanine aminotransferase (U/L) | 15 | (14–20) | ||
| eGFR (mL min–1 1.73 m–2) | 80.8 | (68.5–93.8) | ||
| Duration of diabetes (years) | 12 | (7–21) | ||
| Type of Type 1 Diabetes (slowly progressive/Acute/Fulminant) | 3/ | 10/ | 0 | |
| Diabetic retinopathy (none/SDR/PPDR/PDR) | 8/ | 2/ | 0/ | 3 |
| Diabetic nephropathy (stage 1/2/3/4) | 11/ | 2/ | 0/ | 0 |
| Diabetic neuropathy (stage 1/2/3/4) | 6/ | 3/ | 3/ | 1 |
Data are presented as n or median (interquartile range)
HDL high density lipoprotein, LDL low density lipoprotein, eGFR estimated glomerular filtration rate, SDR simple diabetic retinopathy, PPDR preproliferative diabetic retinopathy, PDR proliferative diabetic retinopathy
Table 2 presents data for the primary and secondary endpoints. The primary endpoint, the change in SG values between before and after breakfast, showed a significant reduction from 90.7 (62.0–116.5) mg/dL before to 51.0 (36.0–62.5) mg/dL after administration of rapid-acting insulin on awakening. The secondary endpoint of the change in SG values between 0300 and 0700 h also manifested a significant decrease from 67.7 (55.2–85.5) mg/dL before to 29.0 (14.3–65.5) mg/dL after such insulin administration. Moreover, SG fluctuations from 0300 to 0900 h showed a corresponding significant decline.
Table 2.
Primary and secondary outcomes on days before and after early-morning administration of rapid-acting insulin
| Outcome | Before therapy | After therapy | P | ||
|---|---|---|---|---|---|
| SG change from 0700 to 0900 h (mg/dL) | 90.7 | (62.0–116.5) | 51.0 | (36.0–62.5) | < 0.01 |
| SG change from 0300 to 0700 h (mg/dL) | 67.7 | (55.2–85.5) | 29.0 | (14.3–65.5) | < 0.01 |
| SG change from 0300 to 0900 h (mg/dL) | 172.5 | (142.8–196.0) | 78.3 | (43.7–97.0) | 0.001 |
| Average blood glucose (mg/dL) | 192.7 | (159.6–210.4) | 156.7 | (134.2–188.0) | 0.03 |
| MAGE (mg/dL) | 164.0 | (135.0–189.9) | 96.1 | (73.6–135.7) | < 0.01 |
| SD (mg/dL) | 64.8 | (60.7–69.1) | 55.9 | (39.3–64.2) | 0.02 |
| CV (%) | 37.8 | (32.7–41.6) | 33.3 | (25.1–38.6) | 0.15 |
| Time above range (> 180 mg/dL) (%) | 48.6 | (35.4–58.0) | 26.9 | (20.8–52.8) | 0.08 |
| Time in range (70–180 mg/dL) (%) | 51.4 | (42.0–57.1) | 61.6 | (42.7–72.2) | 0.13 |
| Time below range (< 70 mg/dL) (%) | 0.0 | (0.0–2.6) | 3.0 | (0.0–7.1) | 0.16 |
| Total insulin dose (U) | 30.0 | (23.2–34.0) | 29.3 | (23.0–31.5) | < 0.01 |
| Basal insulin dose (U) | 7.0 | (4.0–9.0) | 6.7 | (4.0–9.0) | 0.37 |
| Bolus insulin dose (U) | 21.0 | (17.2–28.0) | 21.0 | (16.5–25.5) | < 0.01 |
Data are presented as median (interquartile range). P values for the differences between before and after therapy were determined with the Wilcoxon signed-rank test
SG sensor glucose, MAGE mean amplitude of glycemic excursions, SD standard deviation, CV coefficient of variation
The daily average SG level improved significantly from 192.7 (159.6–210.4) mg/dL before to 156.7 (134.2–188.0) mg/dL after early-morning insulin administration. In addition, there was a significant improvement in MAGE, which decreased from 164.0 (135.0–189.9) mg/dL to 96.1 (73.6–135.7) mg/dL. A plot of the average SG level before and after such insulin administration reveals the suppression of the dawn phenomenon after the intervention (Fig. 2). The total insulin dose decreased significantly from 30.0 (23.2–34.0) units before to 29.3 (23.0–31.5) units after the intervention (Table 2). Whereas the basal insulin dose did not show a significant change, the bolus insulin dose was decreased significantly. When expressed as mean ± standard deviation, the data are as follows: the total insulin dose changed from 30.3 ± 9.1 to 29.1 ± 9.0 (U), the basal insulin dose from 7.6 ± 5.0 to 7.5 ± 5.0 (U), and the bolus insulin dose from 22.6 ± 6.0 to 21.6 ± 5.9 (U).
Fig. 2.
Sensor glucose variation before and after early-morning administration of rapid-acting insulin. Data are the sensor glucose values at each time point averaged for the 13 study individuals
Discussion
Our study has shown that administration of a small dose of rapid-acting insulin on awakening is able to mitigate postbreakfast hyperglycemia in individuals with T1D who experience the dawn phenomenon. Individuals with T1D who rely on frequent insulin injections often face the dilemma of accepting hypoglycemia as a consequence of an increased basal insulin dose or enduring the dawn phenomenon without adjustment of basal insulin. Insulin pump therapy has therefore been advocated to manage the dawn phenomenon and subsequent postprandial blood glucose elevation for persons with T1D [8]. Such therapy offers the advantage of being able to adjust the insulin infusion rate according to diurnal fluctuations in insulin demand. Adjustment of basal insulin levels has revealed that insulin requirements rise after 0300 h [11–13]. Increased basal insulin delivery around 0300 h therefore allows control over the dawn phenomenon and postbreakfast hyperglycemia. However, some individuals with T1D prefer MDI therapy or face barriers imposed by procedural or cost considerations, preventing them from opting for an insulin pump. One known method for addressing the dawn phenomenon involves modifying the formulation of basal insulin, as well as adjusting the timing and dosage of its administration. However, its effectiveness has not been clearly established. So, it is important that the present study demonstrates the marked effectiveness of administration of rapid-acting insulin on awakening with regard to management of the dawn phenomenon and the associated postbreakfast hyperglycemia for such people.
Blood glucose fluctuations have been found to adversely affect vascular endothelial function and oxidative stress [14]. Our results now reveal that early-morning administration of rapid-acting insulin also improves indices of diurnal blood glucose fluctuation, such as MAGE and SD, potentially leading to the prevention or amelioration of vascular endothelial dysfunction and oxidative stress. We also found that such treatment resulted in a significant reduction in the daily total insulin dose, associated with a lowering of average blood glucose levels, which can be explained by the suppression of postbreakfast hyperglycemia after the dawn phenomenon and a consequent reduced requirement for corrective insulin.
A strength of our study is its execution in a controlled hospital setting, ensuring uniformity in dietary energy intake, inclusive of carbohydrate content, and a standardized bedtime regimen. Furthermore, the use of CGM facilitated the consistent monitoring of blood glucose levels from midnight to 0300 h, thereby bolstering the validity of the study. However, the generalizability of our findings is constrained by a limited participant pool and the variability in the timing of insulin administration across individuals. Nevertheless, the influence of external environmental factors is substantially reduced by centering the research on hospitalized individuals. Scheiner et al. indicated that the early morning to midnight basal rate was higher in individuals aged 61 and older compared to other age groups [11]. Given that the median age in our study was 68, this implies that our study population likely included a higher proportion of older individuals who may have been more susceptible to a pronounced dawn phenomenon. In conclusion, our findings affirm that the early-morning administration of rapid-acting insulin is an effective strategy to manage postbreakfast hyperglycemia subsequent to the dawn phenomenon, and therefore represents a viable alternative therapeutic avenue for individuals with T1D who are unable or opt not to utilize insulin pump therapy.
Supplementary Information
Below is the link to the electronic supplementary material.
Author contributions
Tomofumi Takayoshi and Yushi Hirota conceived the study, analyzed the data, and wrote the manuscript. Akane Yamamoto, Kai Yoshimura, Seiji Nishikage, Mariko Ueda and Wataru Ogawa contributed to discussion and interpretation of the data. Wataru Ogawa contributed to revision of the manuscript. All authors read and approved the final manuscript.
Data availability
The data sets generated during the current study are available from the corresponding author upon reasonable request.
Declarations
Conflict of interest
Yushi Hirota has received lecture fees from Eli Lilly Japan K.K., Sanofi Aventis, Abbott Japan, Terumo Co., and Sumitomo Pharma Co. Ltd.; research funding from Sumitomo Pharma Co. Ltd., Medtronic Japan Co. Ltd., and Kyowa Kirin Co. Ltd.; and a donation from Abbott Japan. Wataru Ogawa has received lecture fees from Sumitomo Pharma Co. Ltd., Nippon Boehringer Ingelheim Co. Ltd., Abbott Japan, and Novo Nordisk Pharma Ltd.; research grants from Noster Inc., Nippon Boehringer Ingelheim Co. Ltd., Eli Lilly Japan K.K., Abbott Diabetes Care UK Ltd., Sumitomo Pharma Co. Ltd., and Teijin Pharma Ltd.; and donations from Kowa Co. Ltd., Novo Nordisk Pharma Ltd., Sumitomo Pharma Co. Ltd., Takeda Pharmaceutical Co. Ltd., and Teijin Pharma Ltd. All remaining authors declare that they have no conflict of interest.
Ethical approval
13 November 2018/ approval no180249.
Informed consent
Informed consent was obtained in the form of an opt-out on the hospital.
Human rights
The ethics committee of Kobe University Graduate School of Medicine.
Animal studies
N/A.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Schmidt MI, Hadji-Georgopoulos A, Rendell M, et al. The dawn phenomenon, an early morning glucose rise: implications for diabetic intraday blood glucose variation. Diabetes Care. 1981;4:579–85. 10.2337/diacare.4.6.579 [DOI] [PubMed] [Google Scholar]
- 2.Rybicka M, Krysiak R, Okopień B. The dawn phenomenon and the Somogyi effect—two phenomena of morning hyperglycaemia. Endokrynol Pol. 2011;62:276–84. [PubMed] [Google Scholar]
- 3.Porcellati F, Lucidi P, Bolli GB, et al. Thirty years of research on the dawn phenomenon: lessons to optimize blood glucose control in diabetes. Diabetes Care. 2013;36:3860–2. 10.2337/dc13-2088 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Monnier L, Colette C, Dunseath GJ, et al. The loss of postprandial glycemic control precedes stepwise deterioration of fasting with worsening diabetes. Diabetes Care. 2007;30:263–9. 10.2337/dc06-1612 [DOI] [PubMed] [Google Scholar]
- 5.Zheng X, Qi Y, Bi L, et al. Effects of exercise on blood glucose and glycemic variability in type 2 diabetic patients with dawn phenomenon. Biomed Res Int. 2020;2020:6408724. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Schiel R. Continuous subcutaneous insulin infusion in patients with diabetes mellitus. Ther Apher Dial. 2003;7:232–7. 10.1046/j.1526-0968.2003.00038.x [DOI] [PubMed] [Google Scholar]
- 7.Ostrovski I, Lovblom LE, Scarr D, et al. Analysis of prevalence, magnitude and timing of the dawn phenomenon in adults and adolescents with type 1 diabetes: descriptive analysis of 2 insulin pump trials. Can J Diabetes. 2020;44:229–35. 10.1016/j.jcjd.2019.08.003 [DOI] [PubMed] [Google Scholar]
- 8.Koivisto VA, Yki-Jarvien H, Helve E, et al. Pathogenesis and prevention of the dawn phenomenon in diabetic patients treated with CSII. Diabetes. 1986;35:78–82. 10.2337/diab.35.1.78 [DOI] [PubMed] [Google Scholar]
- 9.Beck RW, Tamborlane WV, Bergenstal RM, et al. The T1D exchange clinic registry. Endocrinol Metab. 2012;97:4383–9. 10.1210/jc.2012-1561 [DOI] [PubMed] [Google Scholar]
- 10.Kanda Y. Investigation of the freely available easy-to-use software ‘EZR’ for medical statistics. Bone Marrow Transplant. 2013;48:452–8. 10.1038/bmt.2012.244 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Scheiner G, Boyer BA. Characteristics of basal insulin requirements by age and gender in Type-1 diabetes patients using insulin pump therapy. Diabetes Res Clin Pract. 2005;69:14–21. 10.1016/j.diabres.2004.11.005 [DOI] [PubMed] [Google Scholar]
- 12.Kuroda A, Kaneto H, Yasuda T, et al. Basal insulin requirement is ~30% of the total daily insulin dose in type 1 diabetic patients who use the insulin pump. Diabetes Care. 2011;34:1089–90. 10.2337/dc10-2149 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Nakamura T, Hirota Y, Hashimoto N, et al. Diurnal variation of carbohydrate insulin ratio in adult type 1 diabetic patients treated with continuous subcutaneous insulin infusion. J Diabetes Invest. 2014;5:48–50. 10.1111/jdi.12132 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ceriello A, Esposito K, Piconi L, et al. Oscillating glucose is more deleterious to endothelial function and oxidative stress than mean glucose in normal and type 2 diabetic patients. Diabetes. 2008;57:1349–54. 10.2337/db08-0063 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data sets generated during the current study are available from the corresponding author upon reasonable request.