INTRODUCTION

Type 1 diabetes mellitus (DM1) represents between 5 to 10% of all diabetes cases and is a condition where beta cells are destroyed by an autoimmune phenomenon. The prevalence and incidence of DM1 are approximately 5.6 and 15 cases per 100,000 inhabitants, respectively, worldwide¹, with the number of new cases in children and young people during the COVID-19 pandemic being alarming².

In Peru, the incidence is low, with approximately one case per 100,000 inhabitants in children under 15 years of age³, but it is increasing. Recently, the National Institute of Child Health in Peru reported a greater number of new cases per year between 2011 and 2018⁴. Patients with DM1 have greater difficulty in reaching glycated hemoglobin (HbA1c) goals and thus be able to reduce complications. Fear of hypoglycemia and poor adherence to treatment in these patients greatly affect good glycemic control. In parallel, hypoglycemia increases mortality and cardiovascular risk⁵.

Self-monitoring is essential for the control of DM1; however, the cost of fingerstick glucose monitoring and the discomfort and pain associated with it limits its use. In developed countries, the use of continuous glucose monitoring (CGM) has become the main technological tool for real-time glycemic control. In Peru, there is a previous report on the use of CGM in an adult⁶and a retrospective study of 28 patients between children and adults⁷.

In this report, we describe the case of a patient with DM1 with poor metabolic control and recurrent hypoglycemia in whom the use of real-time CGM reduced hypoglycemic episodes, improved quality of life, and reached the appropriate dose of insulin.

DESCRIPTION OF THE CLINICAL CASE

A 39-year-old female patient from the northern region of Peru, with a personal history of two previous abortions and DM1 since she was 23 years of age, with a report of an episode of diabetic ketoacidosis currently undergoing treatment with insulin therapy. On the first visit, the patient is evaluated by teleconsultation due to confinement due to the COVID-19 pandemic. Capillary glucose was measured 1 or 2 times a day, 50-70% of the time with a value less than 70 mg/dl, several of the episodes were asymptomatic. She suffered from almost daily episodes of hypoglycemia either day and/or night, with documented symptoms up to approximately 35mg/dl of capillary glucose, some of them were severe nocturnal hypoglycemia that required the assistance of another family member with glucose reaching up to 23 mg/dl.

He has a history of an arm fracture due to severe hypoglycemia with loss of consciousness. His height is 163 cm and weight 49 Kg (BMI 18.4 Kg/m2), body mass index is usually less than 20 Kg/m2, he used NPH and Regular insulin with a total daily dose of insulin (TDD) 1 at 1.2 IU/KG (seeTable 2). Auxiliary tests showed the following results: HbA1c 8.1%, C-peptide <0.01 nmol/L, total cholesterol: 197 mg/dl, high-density lipoprotein (HDL) 56 mg/dl, low-density lipoprotein (LDL) 138 mg/dl, triglycerides: 65mg/dl, thyroid-stimulating hormone (TSH): 1.52mIU/L, creatinine 1.44mg/dl, and microalbuminuria 1.4mg/L.

Six days after the teleconsultation, a Free Style Libre 2 CGM sensor was placed. Given the high dose of daily insulin and the high frequency of hypoglycemia, the change to insulin analogs was made according to weight: glargine 18 IU/day and lispro 4/7/5 IU before breakfast, lunch, and dinner, respectively, while providing nutritional education and counseling (seeTable 2). Following the manufacturer's instructions, the sensor was applied to the skin of the outer aspect of the arm. Likewise, the Libre 2 scanner was provided and he was instructed to scan the CGM sensor at least every 8 hours during the 14 days that the sensor lasts. The patient sent daily insulin dose information and photos of her food. Correction of insulin doses was coordinated with her and the accuracy of carbohydrate counting was reinforced. Additionally, during the 2 weeks of CGM use, the patient made her own minimal adjustments to the basal and preprandial insulin doses, empowering her in the education and management of her diabetes.

The report of the first CGM obtained a time in the range of 70 to 180 mg/dl in 66% (the hypoglycemia alarm setting was from 70 mg/dl), an average glucose of 134 mg/dl, a glucose variability 46.2% and 26 low glucose events with an average duration of 120 minutes (Figure 1). The basal insulin dose was 18 IU/day and preprandial insulin was finally divided into 2/2/2 IU on average (Table 2).

At the four-month outpatient follow-up, she showed better control of self-monitoring of glucose, with capillary glucose measurement four times a day, and fewer hypoglycemia episodes. It was decided to install another CGM for two weeks, with an alarm setting from 80mg/dl to prevent hypoglycemic episodes earlier. An average glucose value of 137 mg/dl, a target range at 70%, a glucose variability of 46.6% and 18 low glucose events with an average duration of 117 minutes were recorded (SeeFigure 1). Control laboratory tests showed the following results: HbA1C: 7.3%, total cholesterol 162 mg/dl, HDL 46.4 mg/dl, LDL 104.9 mg/dl, triglycerides 63.8 mg/dl, creatinine 0.4 mg/dl, TGP or Alanine Aminotransferase 19 mIU/mL.

Figure 1 Report of glucose statistics and time in range using Continuous Glucose Monitor MCG: FreeStyle Free 2 Continuous Glucose Monitor 

The dose at four months of basal insulin with insulin glargine was 21 IU/day and insulin lispro 3/3/3 IU/day. The patient reports less hypoglycemia and better quality of life, in addition to a better understanding of the effect of the number of carbohydrates, which allows her to have more control over the disease.

DISCUSSION

Hypoglycemia causes cognitive impairment and is associated with mortality in 4-10% of patients with DM1, and cases have been described of patients dying during sleep after long periods of asymptomatic hypoglycemia^9,10. In the present report, our patient decreased the number of hypoglycemic events between the first and second CGM from 24 to 18 episodes, respectively. Although the average duration was similar (120 minutes versus 117 minutes), the total time of hypoglycemia in the 14 days that each GCM lasted decreased from 3120 minutes (3.8 hours per day) in the first GCM to 2106 minutes (2.5 hours per day) in the second GCM. The higher the frequency of hypoglycemia, the higher the reactive hyperglycemia due to the fear of a hypoglycemic episode⁹. Recently, international guidelines, in addition to glycosylated hemoglobin, recommend goals based on the GCS as the longest time in range, the shortest time in hypoglycemia and the lowest glycemic variability to mitigate long-term complications¹¹.

In the case of our patient, she reported a decrease in hypoglycemia episodes with the switch to insulin analogs. This suggests that, with the previous regimen of NPH and regular with which she had higher daily insulin doses, the exposure to daily hypoglycemia was probably even worse. The use of real-time CGM allowed the patient to better understand the impact of carbohydrate intake and its relationship to insulin doses.

The availability of hypoglycemia alarms on the MCG is especially useful in patients with a history of asymptomatic hypoglycemia as in our patient. Many of the episodes of hypoglycemia are nocturnal and the alarm shortens the time of exposure to hypoglycemia, as was observed in a pediatric patient who did not react to hypoglycemia when using a CGM without a hypoglycemia alert, but did when she switched to a CGM with alerts that helped reduce the duration of hypoglycemia¹².

Of special interest is the patient's lack of reaction to the MCG hypoglycemia alert as seen inFigure 2. The upper panel shows 20 scans (white circles), two of them with hypoglycemia around 3 PM. The previous 15 hours there was no hypoglycemia. The glucose level is seen to rise 20 minutes after hypoglycemia was detected. The next day the patient had a 2-hour episode of hypoglycemia between 2:20 and 4:40 AM. She scanned the MCG at 4:30 which yielded a value of 52 and within minutes her glucose level began to rise indicating carbohydrate intake. However, in the afternoon of the same day she had a prolonged episode of hypoglycemia between 1pm and 5:30pm. The patient performed several scans approximately 1 per hour, several of them corresponding to hypoglycemia, but the glucose level remained at about the same level suggesting no carbohydrate intake until 5:30pm, when she developed marked reactive hyperglycemia, greater than 300 mg/dL for several hours, suggesting excessive carbohydrate intake.

Hypoglycemia is known to transiently and sometimes permanently affect cognitive function¹³, it is possible that the frequent episodes of hypoglycemia, especially that of the same day, may have affected the patient's decision-making ability, predisposing her to develop more hypoglycemia that same day in the afternoon.

Figure 2 Sample Daily Record of Glucose Levels * MCG: FreeStyle Free 2 Continuous Glucose Monitor 

In a previous report, the case of a 29-year-old patient with long-standing DM1 who used a blinded CGM (Lizarzaburu) was described. The blinded CGM does not show the patient the glucose level in real time but at the end of the period of use (6 to 14 days later), in the mentioned case, the CGM showed a time in target range of 17% and persistent hyperglycemia, it is mentioned that there was lack of adherence to nutritional control and physical activity; subsequent controls at the third and fifth year showed gradually less hyperglycemia, better glycemic control (HbA1c) and with little evidence of hypoglycemia, however, she reached a higher BMI than at the beginning.

To our knowledge, the present report is the first Peruvian case of improvement of severe hypoglycemia and decrease in the frequency of asymptomatic hypoglycemia with the use of CGM in DM1. In addition to visualizing her glucose level, the patient had a very close medical and nutritional follow-up, which enhanced nutritional education and empowerment. This improvement continued after 4 months of follow-up when she even increased the time in target range from 66% during the first CGM to 70% in the second CGM, in addition to a clinically significant reduction of 1.3 hours of hypoglycemia per day.

In our setting, a pilot study of 28 patients in pediatric and adult populations showed that the use of the CGM, although for a short period of time, was beneficial in either type 1 diabetes (predominantly children) or type 2 diabetes (predominantly adults). In that study, those who used the Free 14-day MCG system with scanning, but without a hypoglycemia alarm, there was a correlation between the number of scans and time in the range 70-180 mg/dl, while adults who used the blinded monitor had no behavioral feedback of carbohydrate intake as the patient was "blind" to the glucose level⁷. 100 years after the discovery of insulin, patients with diabetes have empowered their diabetes treatment thanks to more stable and predictable insulins in their absorption (analogs) and the GCM technology that allows an approach to the ideal therapy of the patient with DM1. In conclusion, the CGM, especially incorporating an alarm system, is a useful tool to consider in patients with diabetes and asymptomatic hypoglycemia as it allows an optimization of the insulinization schedule therapy, empowerment of the patient with carbohydrate intake and insulin use, and a clinically significant decrease in the frequency and total duration of hypoglycemia.