Glycaemic outcomes of a 4‐h insulin pump suspension for prolonged exercise in youth with type 1 diabetes

Managing prolonged exercise is a significant challenge for youth with type 1 diabetes using continuous subcutaneous insulin infusion (CSII), particularly when an activity requires pump disconnection for over an hour.1, 2 During and after prolonged exercise (over 75 min), the reduced or absent basal insulin in individuals with diabetes increases the risk of both hypoglycaemia and subsequent rebound hyperglycaemia. While laboratory-based studies provide valuable insights into metabolic responses to structured exercise, there is a paucity of evidence guiding strategies for unpredictable, real-world activities that necessitate pump removal. Management strategies include consuming readily absorbable carbohydrates to prevent low blood sugar and potentially reducing basal insulin delivery before or during activity. Frequent blood glucose monitoring, especially using continuous glucose monitors (CGM), is crucial to manage these risks effectively, but limited evidence exists to guide specific strategies for very long durations of exercise.3 We report the glycaemic outcomes of a pragmatic strategy used during a 4-h sailing activity in a real-world setting. Capillary blood glucose was measured before breakfast, pre-exercise, postexercise and 4 h after exercise completion (5:00 PM). A total of 186 valid glucose measurements were collected across four separate sailing activities during the camp. A repeated measures ANOVA revealed a significant effect of time on glycaemic values (p < 0.01). As shown in Table S1, mean glucose levels rose from 8.6 ± 3.9 mmol/L (155 ± 71 mg/dL) at pre-exercise to a peak of 9.1 ± 3.4 mmol/L (164 ± 61 mg/dL) immediately after exercise, before falling to 7.2 ± 3.6 mmol/L (129 ± 65 mg/dL) 4 h later. Crucially, a clinical risk analysis (Figure 1) revealed the trade-offs of this strategy. The protocol was highly effective at preventing hypoglycaemia during exercise (only 2.1% of postexercise readings were <3.9 mmol/L, <70 mg/dL). However, this safety came at the cost of significant hyperglycaemia, with 36.2% of pre- and postexercise readings exceeding 10 mmol/L, 180 mg/dL. Furthermore, the risk of hypoglycaemia resurfaced in the late postexercise period, with 9.1% of readings being hypoglycaemic 4 h after the activity, a known phenomenon of delayed-onset hypoglycaemia.4, 5 The pronounced hyperglycaemia observed pre- and postexercise likely resulted from the cumulative effect of the reduced breakfast bolus and the prolonged basal insulin deficit, which the mid-activity bolus could not fully overcome. This highlights a key physiological challenge: manually replacing missed basal insulin without causing exercise-induced hypoglycaemia is exceptionally difficult. Alternative approaches, such as significant temporary basal rate reductions prior to disconnection or the use of an untethered insulin patch pump for basal delivery, may warrant investigation.6 This study's strength lies in its real-world evaluation of a practical challenge. However, several limitations must be considered. First, the observational, single-arm design lacks a control group, such as a comparator day without pump disconnection, which would have provided a clearer benchmark for the glycaemic effects of this strategy. Second, the reliance on capillary glucose monitoring, while pragmatic for the setting and the gold standard for therapeutic decisions at the time the study was conducted, likely underestimated the true glycaemic variability and missed asymptomatic excursions that continuous glucose monitoring (CGM) would have captured together with trend data (i.e. rate-of-change arrows), which is a key tool for proactive management during and after exercise, allowing for earlier intervention before hypoglycaemia or significant hyperglycaemia occurs. Third, the exercise intensity of sailing is inherently variable and was not quantified, which could impact glucose utilization differently among participants and across days. Finally, the small sample size limits the generalizability of our findings. These results highlight the delicate balance in managing prolonged activity without advanced AID systems, which can autonomously modulate insulin delivery. In conclusion, our findings demonstrate that for prolonged exercise requiring pump disconnection, this multifaceted strategy can successfully prevent acute hypoglycaemia. However, it fails to adequately control hyperglycaemia and may increase the risk of delayed hypoglycaemia. This creates a crucial benchmark for managing patients not using or having access to advanced AID systems. Modern AID algorithms with exercise functions are specifically designed to mitigate these wide glycaemic excursions by dynamically adjusting insulin delivery.7 While this protocol serves as a safe starting point for non-AID users, it underscores the critical need for individualized adjustments and highlights the potential benefits of CGM and AID systems in safely managing these challenging scenarios. Andrea E. Scaramuzza conceptualized the study, developed the methodology and supervised the project. Andrea E. Scaramuzza was involved in data curation and formal analysis. Riccardo Bonfanti and Ivana Rabbone contributed to the investigation and data discussion. Andrea E. Scaramuzza wrote the original draft. All authors reviewed, discussed and edited the manuscript and approved the final version. The authors thank the study participants and their families for their dedication and commitment. This research did not receive any specific grant from funding agencies. None. All authors declare no conflicts of interest related to the present study. Andrea E. Scaramuzza is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Data S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.