Introduction

Fasting is a ritual that dates back to “Antiquity”. Fasting during “Ramadan”, the ninth lunar month of the Islamic calendar, is currently the most famous and the most practised and consists of abstinence from food and drink from dawn to sunset, which can reach 21 h and more in Scandinavian countries. Prepubescent children and patients with several types of acute or chronic medical conditions are exempted from Ramadan fasting (RF). However, at least 50 million individuals with diabetes around the world practice RF every year. A recent survey indicated that almost four-fifths of physicians treating children and adolescents with diabetes would allow their patients to fast if they wished, provided they have reasonable diabetes control and are well educated against hypoglycaemia [1]. One of the established treatment options for this category is the insulin pump, at least at the theoretical level, but data during RF are limited. Few studies have reported encouraging results with continuous subcutaneous insulin infusion (CSII) therapy to reduce hypoglycaemia and improve glycaemic variability during RF [2]. Petrovski et al. reported a unique case of a teenager fasting safely during Ramadan on a hybrid closed-loop system [3]. The MiniMed 780G, an advanced hybrid closed-loop (AHCL) system (Medtronic, USA), which delivers an automated base in response to glucose sensor readings transmitted to the insulin pump every 5 min, is the latest technology in terms of closed-loop systems and uses an algorithm that could give an automated auto correction bolus when necessary. We present the case of an adolescent with type 1 diabetes (T1D) in whom glucose control was managed using sensor-augmented pump therapy with predictive low-glucose management (SAP-PGLM) MiniMed 640G during the month of Ramadan in 2021 and then the MiniMed 780G AHCL system during the following Ramadan season in 2022. To the best of our knowledge, this is the first patient to report fasting during Ramadan under the AHCL system and the first study to compare the two technologies during RF.

Case presentation

An 11-year-old child with T1D fasted for two successive Ramadan seasons under two different insulin administration technologies that are innovative in terms of diabetes management. He had an 8-year history of diabetes, and his case summarizes the progression from a conventional regimen to an advanced hybrid closed-loop system over the past decade in terms of diabetes management and treatment and the testing of multiple daily injections (MDIs) of insulin and CSII therapy. His diabetes was revealed by DKA at the age of 2 years. With the resolution of DKA, he was put on a conventional insulin regimen. Then, faced with recurrent hypoglycaemia, he was switched to multiple daily injections (MDIs) with blood glucose self-monitoring. At that time, the sensor was not available in the KSA, and his HbA1c levels were approximately 8% (63.9 mmol/mol). He was among the first patients in KSA to use the insulin pump: The MiniMed Paradigm VEO. This pump involves an open-loop CSII system, which was initiated 6 years ago for severe nocturnal hypoglycaemia with MDIs and led to improved glycaemic control in terms of hypoglycaemia. Then, he was started on the MiniMed 640G with the algorithm predictive low-glucose management and an HbA1c level of 7.5% (58.5 mmol/mol) and a 52% time-in-range (TIR) (70–180 mg/dl; 3.9–10.0 mmol/l) was reached. The patient was referred from our branch in the Joubail Pediatric Department for our paediatric endocrinology consultation on Dammam 2 years ago, and an HbA1c level of 6.8% (50.8 mmol/mol) and a TIR (70–180 mg/dl; 3.9–10.0 mmol/l) of 60% were achieved. He was not advised to fast, as he did not fulfil the ISPAD 2020 criteria for fasting, but he was identified via our pump patient education WhatsApp group FR. He caused us to accept reality, and he was advised to decrease the basal rate by 20% post afternoon and increase the meal bolus by 20% for Iftar and to reduce it by 10% for Suhoor.

The MiniMed 780G AHCL system was suggested for the patient for uncontrolled down phenomenon and initiated 4 months before Ramadan 2022. An HbA1c level of 6.6% (49 mmol/mol) and a time-in-range (70–180 mg/dl; 3.9–10.0 mmol/l) of 71% was reached.

The patient and his father were included in the watts app Mouwast 780G group, which supports patients on a closed-loop insulin pump from the Mouwast group. This group is dedicated to interactive complementary education. Our patient fulfilled the ISPAD criteria for fasting in 2022 and received pre-Ramadan education through this watts app group, in which patients were instructed to break their fast if their blood sugar level reached < 70 mg/dl or > 300 mg/dl or if they experienced symptoms of hypoglycaemia. We advised all patients to use temporary targets 4 h before sunset and if their glucose levels reached 80 mg/dl (4.4 mmol/l) before sunset to avoid further glucose decrease. If unexplained hypoglycaemia occurred, a temporary target was used 2 h before the onset time of the hypoglycaemic episode. Diet education was also given through the watts app group, and from our previous experience with 670 G in the previous Ramadan season (data not yet published), all patients were advised to increase the meal bolus by 10–20% if the meal contained more than 100 g and to split the bolus insulin to 70% before and 30% after the meal to reverse insulin resistance and to cover high specific fat content, as automated auto correction alone cannot cover some traditional foods with high fat levels that are specific to Ramadan.

We extracted data on glycemic control from reports generated with CareLink™ software for healthcare professionals, and consultation was given over the phone during the first week and face-to-face during mid-Ramadan. Glucose and insulin metrics were analysed 1 month before and during the Ramadan seasons. The patient had a sensor availability problem for 1 week 1 month before Ramadan (1MBR) in 2021 and 2022. For comparison, January 2021 was considered the month before the Ramadan 2021 season, and February 2022 was considered the month before the Ramadan 2022 season, as there was no significant difference in sensor wear (SW) and smart guard (SG) per week between Ramadan and these 2 months and the setting was the same as that during Ramadan (Fig. 1).

Fig. 1
figure 1

Glucose control during Ramadan and the time-in-range glucose distribution: 24-h glucose overlay, percentile comparison of the two periods; an advanced hybrid closed-loop system (MiniMed 780 G) was used during Ramadan fasting in 2022 (blue colour) and a sensor-augmented pump therapy with predictive low-glucose management (Minimed 640G) was used during Ramadan fasting in 2021 (grey colour). 1MBR: one month before Ramadan, SG: SmartGuard (i.e., automated basal insulin + autocorrection of Medtronic insulin pump technology), SW: sensor wear, GMI: glucose management indicator, NA: not applicable, p: significance

Full size image

The patient broke the fast 6 times with the SAP-PGLM system and three times in the second Ramadan with the AHCL system due to mild hypoglycaemia. Despite correct carbohydrate counting, the patient was unable to control his post-Iftar blood glucose(rising 2 h after the meal), and a slight increase in glucose values (20–00 h) was noted due to the fat and protein effect of the Iftar meal (as shown in Fig. 1). The AHCL system and automated self-correction alone did not cover some traditional high-fat foods that are specific to Ramadan. Supplementation with 15% extra carbohydrates was advised specifically for our patient 1 h and 30 min after Iftar to control this phenomenon with good results. However, we faced a problem because the child participated in group prayer 1 h and 30 min after the meal, and he refused to publicly manipulate the pump to deliver the extra bolus to anticipate a hyperglycaemia peak 2 h post-Iftar. This extra bolus was given with a 30-min delay coinciding with the peak. On the other hand, a slight modification of the carbohydrate ration by 15%, associated with fixation of the target at 120 mg/dl, was able to control this phenomenon; however, the child did not tolerate this strict control well and experienced false feelings of hypoglycaemia.

Data and details of glycemic control and insulin metrics on SAP-PGLM and AHCL during R and 1MBR were summarized (Table 1 and Fig. 1) and analysed with statistical methods by SPSS. We did not find any significant difference in the total insulin per day before and during Ramadan with SAP-PGLM or with AHCL (as shown in Table 1). However, glycemic control was significantly better during RF with AHCL than during RF with SAP-PGLM (Table 1). The comparison of the time spent in range and paradoxically spent in hypoglycaemia is in favour of best control during RF with AHCL (Fig. 1) and no diabetic Ketoacidosis or severe hypoglycaemia were detected during and before Ramadan.

Table 1 Statistics during Ramadan vs. 1 month before Ramadan for continuous subcutaneous insulin infusion through open-loop sensor-augmented pump therapy with predictive low-glucose management (SAP-PGLM) vs. an advanced hybrid closed-loop (AHCL) system
Full size table

Discussion

In 2021, through their telemonitoring experience, Bedowra Zabeen et al. reported that young people with type 1 diabetes on CSII insulin pump therapy using the stop before low algorithm fast safely during the COVID-19 pandemic Ramadan [2]. Despite certain limitations and younger age, we share the same findings reported by Zabeen et al. There were episodes of hyperglycaemia and mild hypoglycaemia, but we have provided a more detailed overview (expressed as % of the time spent in each event) of glucose control in a child with T1D during the month of Ramadan. We did not find any significant difference in the total insulin per day before and during Ramadan (as shown in Table 1), and the CSII system with minimal setting changes seems to show results that are not influenced by RF in terms of time spent in hypoglycaemia (Fig. 1). Previous encouraging results with the low-glucose suspend algorithm were approved by the ISPAD in 2020. To the best of our knowledge, the new advancement of the insulin pump, the advanced hybrid closed-loop system, has never been reported in adolescents during RF; however, it seems to be more effective in terms of glycaemic control in children over 7 years via clinical trials. The first-generation hybrid closed-loop (MINIMED Medtronic 670G) system without an autocorrecting algorithm was reported in a unique experience during RF, and no changes in TIR were reported [3]. The same device has been successfully used in different challenging experiments and led to optimal perioperative glycaemia in a 9-year-old child with T1D [4]. We present the case of a T1D patient in whom glucose control was managed using the AHCL system during 1 month of continuous diurnal fasting.

To the best of our knowledge, our case presents the first opportunity to compare the SAP-PGLM to the automated insulin delivery system with autocorrection during FR. The superiority of the AHCL system is illustrated in Fig. 1. This superiority in glycaemic control seems to be related to a higher total daily insulin dose (P < 0.05) and a significantly higher percentage of bolus amount per day in R compared to to the SAP-PGLM (p < 0.001) for the same ICR (insulin carb ratio) (Table 1). It, therefore, reflects the effect of autobasal and autocorrection to maintain glycaemic control in the therapeutic target during the 14 h of continuous fasting and counteracts the insulin resistance of pre-Iftar meal (Fig. 1) with better safety (less glycaemic variability, approximately 6.1%). Despite certain limitations with such a new generation of insulin pumps, our patient showed satisfactory glucose control without a significant change in the TIR and the time spent in hyperglycaemia during Ramadan (critical situation) and 1 MBR (noncritical situation) using the AHCL system. The paradoxically significant less time spent in hypoglycaemia could be related to more pre-Ramadan education focused on hypoglycaemia and the hasty preventative use of temporary targets to avoid breaking fasts.

Automatic basal adjustments and autocorrection with the MiniMed 780G are more practical and safer during Ramadan [3, 5] and seem to be affected by the autocorrection algorithm specific to the MiniMed 780G. Similar to several other studies, we did not diagnose any diabetic ketoacidosis or severe hypoglycaemia during the month of Ramadan in either the SAP-PGLM or HCL, as reported in adults [2].

The AHCL system, with automatic adjustments of basal insulin delivery and automated bolus correction in response to continuous glucose monitoring readings and a flexible temporary target feature, allows improved glycaemic outcomes and reduced hypoglycaemia during fasting periods of more than 14 h per day during the month of Ramadan, which we cannot be reached using the advanced CSII with suspend before low system. Further study on a larger scale should be performed to confirm our findings.