Time in Tight Range Across Menstrual Cycle Phases in Women with Type 1 Diabetes Using the MiniMed 780G System: From the 780MENS Study

To the Editor,

Fluctuations in glycemic control across the menstrual cycle constitute a relevant clinical problem often reported by women with type 1 diabetes (T1D). Specifically, increases in blood glucose levels are often observed during the late luteal phase compared with the early follicular phase. Interestingly, some recent publications have evaluated the performance and ability of different advanced hybrid closed-loop (AHCL) systems to mitigate these glycemic variations throughout the menstrual cycle in women with T1D. Although evidence is still scarce, most previous studies found no differences across phases according to several glycemic metrics, including time in range (TIR, 70–180 mg/dL), a key parameter in continuous glucose monitoring (CGM).^1–3 In this regard, while no differences in TIR were detected between phases in our previous work in women using the MiniMed™ 780G AHCL system, a statistically significant increase in average glucose, glucose management indicator, and time above range (>250 mg/dL) was observed in the late luteal phase. ¹ Significant increases in insulin delivery in the late luteal phase were also detected, suggesting that the system was only partially able to mitigate higher glucose levels in this phase. ¹ In contrast, Elhenawy et al. demonstrated that the MiniMed 780G AHCL system eliminated differences between phases for all CGM metrics, with a significant improvement as compared with the open-loop period. ² In addition, Levy et al. described phase-related changes in group-level TIR and average glucose using the Tandem Control-IQ AHCL system, although no intra-participant differences were detected between phases and no formal statistical analyses were performed. ³ However, it is noteworthy that, to our knowledge, no previous works have evaluated potential differences across the menstrual cycle regarding time in tight range (TITR, 70–140 mg/dL), an interesting novel metric that might have relevant clinical implications.

Recently, we conducted and reported results from an observational study involving 12 premenopausal participants with T1D who were users of the MiniMed 780G AHCL system, in which 36 consecutive menstrual cycles were prospectively recorded and analyzed. ¹ Notably, we observed similar results between the late luteal phase (the last 7 days preceding the onset of menstruation) and the early follicular phase (the first 7 days starting from the first day of menstruation) for most glycemic metrics, including TIR (83.0 [76, 87.5] vs. 85.0 [79.8, 89.0] %—median and interquartile range, respectively; P = 0.101). However, we reported a higher average glucose in the late luteal phase compared with the early follicular phase (139.5 [133.5, 145.2] vs. 131.5 [126.8, 140.2] mg/dL, respectively; P = 0.002). Significant differences in total insulin dose, which was higher in the late luteal phase, were also detected.

In this subanalysis of the 780MENS study, we aimed to focus on TITR throughout these two phases of the menstrual cycle. The methodology of this study has been previously detailed, and all the study participants and menstrual cycles were included in the analyses, which were performed using linear mixed models as described previously. ¹ A TITR of 60.8 ± 9.4% (mean ± standard deviation) was observed during the early follicular phase, whereas the TITR during the late luteal phase was 55.2 ± 8.4%. Notably, these differences were statistically significant (−5.62, 95% confidence interval [CI] −9.28 to −1.96; P = 0.003). In parallel, TIR 140–180 mg/dL was also different between phases, 23.5 ± 5.7% for the early follicular phase, and 26.9 ± 5.9% for the late luteal phase (3.07, 95% CI 0.13–6.01; P = 0.041).

Our main findings reveal that significant differences in TITR may occur throughout the menstrual cycle in women with T1D, even with the use of the MiniMed 780G AHCL system. These results highlight that TITR may offer greater sensitivity than TIR in detecting subtle glycemic variations across menstrual phases, particularly in well-controlled individuals using AHCL systems. This aligns with findings from Bahillo-Curieses et al., ⁴ who showed that TITR was more adequate than TIR in distinguishing levels of glycemic control in both pediatric and adult populations using the MiniMed 780G AHCL system, even when overall control was already within target.

Similar to Elhenawy et al., ² we did not find differences in TIR in the previous analysis of 780MENS study. ¹ Nevertheless, whereas Elhenawy and colleagues reported that the MiniMed 780G AHCL eliminated glycemic differences between menstrual phases when compared with open-loop use, our study identified a significant reduction in other CGM metrics, including TITR, during the luteal phase despite high AHCL system adherence, suggesting an incomplete mitigation of glycemic differences between phases. However, it should be noted that Elhenawy et al. did not evaluate TITR, focusing instead on TIR and other CGM metrics. Moreover, although both studies addressed menstrual cycle-related glycemic variability, our design focused specifically on comparing two clearly defined menstrual phases under consistent AHCL use (>95% auto mode), which may enhance sensitivity to subtle changes. In contrast, Elhenawy et al. compared two treatment periods (open-loop vs. AHCL), which, while also evaluating menstrual phases, may introduce more variability and make subtle differences harder to detect.

In our analysis, a slight but significant increase in TIR 140–180 mg/dL during the luteal phase was observed, which coincided with a decline in TITR, suggesting a redistribution of glucose values toward the upper end of the target range. However, previous studies in different populations demonstrated that the improvement in TITR resulted from the reduction in time above range without changing TIR 140–180 mg/dL. ⁵

Several considerations should be taken into account when interpreting these results. First, although TITR has emerged as a promising metric, its clinical implications are yet to be fully elucidated in adults with T1D. ⁶ Second, while statistically significant, the clinical relevance of the reported between-phase difference could be open for discussion. Indeed, physiological glycemic fluctuations across the menstrual cycle have also been reported in healthy women. ⁷ Therefore, the goal of AHCL systems might be to minimize glycemic fluctuations between menstrual phases to levels comparable with individuals without diabetes rather than striving for the complete absence of differences. Finally, this study has some limitations. Due to the relatively small sample size, our results should be interpreted with caution. Given the excellent glycemic control of the women included in this study, the generalizability of these results may be limited. As in most previous studies, hormonal determinations were not evaluated. In addition, this was a secondary analysis from a previous study, and further investigation is needed to confirm our results. However, several strengths should also be highlighted, including the prospective method of reporting and evaluation of menstrual cycles and careful control for confounders. Furthermore, to our knowledge, this is the first study that evaluates differences in TITR across the menstrual cycle in women with T1D using AHCL systems.

In conclusion, in a cohort of women with T1D using the MiniMed 780G AHCL system, significant differences in TITR were detected between the early follicular and late luteal phases of the menstrual cycle. Further research is warranted to evaluate the performance of additional AHCL systems in relation to this outcome and its clinical implications.

Authors’ Contributions

G.M., M.J.P.-C., and J.I.M.-M.: Conceptualization and study design. G.M. and M.J.P.-C: Data collection. G.M., M.J.P.-C., and J.I.M.-M.: Data analysis and interpretation. G.M., M.J.P.-C., and J.I.M.-M.: Investigation. J.I.M-M.: Original draft preparation. G.M., M.J.P.-C., J.G.-A., F.J.T., and J.I.M.-M.: Article review and editing. G.M., M.J.P.-C., J.G.-A., F.J.T., and J.I.M.-M.: Visualization. All authors read and approved the final version of this article.