Insulin Delivery and Glucose Variability Throughout the Menstrual Cycle on Closed Loop Control for Women with Type 1 Diabetes

Abstract

Objective:

To analyze insulin delivery and glycemic metrics throughout the menstrual cycle for women with type 1 diabetes using closed loop control (CLC) insulin delivery.

Methods:

Menstruating women using a CLC system in a clinical trial were invited to record their menstrual cycles through a cycle-tracking application. Sixteen participants provided data for this secondary analysis over three or more complete cycles. Insulin delivery and continuous glucose monitoring (CGM) data were analyzed in relation to reported cycle phases.

Results:

Insulin delivery and CGM metrics remained consistent during cycle phases. Intraparticipant variability of CGM metrics and weight-based insulin delivery did not change through cycle phases.

Conclusions:

For this sample of menstruating women with type 1 diabetes using a CLC system, insulin delivery and glycemic metrics remained stable throughout menstrual cycle phases. Additional studies in this population are needed, particularly among women who report variable glycemic control during their cycles.

Trial Registration:

NCT03591354.

Introduction

Many women with type 1 diabetes mellitus (T1D) report difficulty with blood glucose management during certain phases of their menstrual cycles. In the first large study evaluating glycemic changes around the menstrual cycle published in 1942, the authors reported that the diagnosis of diabetic ketoacidosis in 36 women previously treated for T1D occurred within 2 days of the menstrual period¹ for about half the women. It is generally agreed upon clinically that premenstrual changes in blood glucose occur in some, but not all, women with T1D, with the percentage of women experiencing a menstrual cycle phenomenon varying from ∼40% to 60%.^2,3

Investigators have used the glucose clamp technique⁴ or the minimal model to quantitate changes more accurately in glucose metabolism during the menstrual cycle in the research setting. Euglycemic clamp studies in women without diabetes have found stable glucose uptake and insulin sensitivity across phases.^5,6 In women with T1D, studies with hyperglycemic hyperinsulinemic clamps by Widom et al. demonstrated that a subgroup of women with T1D who exhibited worsened premenstrual hyperglycemia had a decline in insulin sensitivity during the luteal phase.⁷

Trout et al. studied intravenous glucose tolerance tests in five women using insulin pumps during each cycle phase⁸ and evaluated insulin sensitivity and glucose effectiveness by minimal model analysis,⁹ without statistically significant differences in insulin sensitivity noted in this small group of participants. Although data on changes in insulin sensitivity evaluated in a clinical research setting have been conflicting, it appears changes may occur in a subset of women with T1D. In addition the impact of prior glycemic control or oral contraceptive use around menses for women with T1D has not been well studied.

Two studies have evaluated glycemic control in the real-world setting across the menstrual cycle and have found variations around menses.¹⁰ The largest study to date by Brown et al. studied 12 women on sensor-augmented pump therapy.¹¹ They reported increasing risk of hyperglycemia during periovulation and early luteal phases and lower insulin sensitivity during the luteal phase. The findings tended to be specific to the individual and consistent across cycles.

To date, little information is available regarding insulin delivery or the glycemic impact of closed loop control (CLC) throughout the menstrual cycle. Whether performance of CLC systems varies in women with menstrual cycle-dependent changes in insulin requirements merits evaluation to ensure that the needs of this subset of people with T1D are being appropriately addressed in their use of CLC. Contraception is often a requirement for women with T1D of child-bearing age participating in clinical studies, so evaluation of freely menstruating women as well as the impact of specific types of contraception is to date understudied.

We previously reported the results of the International Diabetes Closed Loop (iDCL) trial, which assessed the efficacy and safety of the Tandem Control-IQ automated-insulin delivery (AID) system in participants 14 years and older during a 6-month randomized controlled trial (RCT)¹² and a ∼12-month extension study.¹³ The purpose of this study is to report on the observed glycemic outcomes of a subset of menstruating women using the Control-IQ system during the iDCL extension study.

Materials and Methods

Study design

Enrolled menstruating women who consented to participate were asked to install a menstrual cycle-tracking app (Clue-BioWink) on their personal smart phones.¹⁴ Participants were provided instructions on how to use the app to report details about their contraception method and menstrual cycle dates in a prospective manner until their final study visit. Enrollment in this substudy could occur at any time convenient for the participant during the parent study; therefore, baseline glycated hemoglobin (HbA1c) was defined as the most recent HbA1c value before the participant's initial use of the cycle-tracking app.

Menstrual phases were defined as contiguous days with vaginal bleeding recorded. Luteal phases were estimated to occur during the 9 days before the onset of menses to ensure an interval of time most consistent with the mid to late luteal phases. The “other” phase was the interval of time between these two phases and would include the follicular phase. Changes in insulin delivery and glycemic responses were evaluated for each cycle phase.

Algorithm description

The CLC system used by all participants during the study consisted of a pump (t:slim X2 insulin pump with Control-IQ, Tandem Diabetes Care) and continuous glucose monitoring (CGM) (Dexcom G6, Dexcom). The CLC algorithm, as has been previously described,¹² has several distinct features: (1) automated insulin delivery through continual basal rate modulation as well as periodic automated correction boluses, (2) attenuation of insulin delivery to avoid hypoglycemia, and (3) gradual lowering of the target to a narrower range overnight. These goals are attained by optimizing insulin on board to achieve desired targets based on predictions of CGM excursions. Although the insulin parameters are preset in a similar manner as a standard pump, the CLC system can vary the insulin infusion rate.

Statistical methods

For this secondary analysis, participant demographics, insulin delivery, and CGM metrics in relation to participants' self-recorded menses from the cycle-tracking app were analyzed. Outcomes were assessed during the mid-late luteal phase, menstrual phase, and rest of the cycle (including the follicular phase) for the entire participant cohort as well as intraparticipant changes in insulin delivery and glycemic responses across cycles. Equal weight was given to each CGM reading within a participant. Given that the study was exploratory and not powered for this secondary analysis, descriptive statistics are provided with no statistical tests to compare groups.

Results

This substudy included 16 women from four study sites, 11 (69%) of whom were White with an age range of 15–45 years (mean 31 ± 12), a duration of diabetes of 5–39 years (median 17 years), median body mass index of 26 kg/m² (interquartile range [IQR] 23–28), and mean HbA1c of 6.8% ± 0.7% (Supplementary Table S1). Participants reported data over an average of 6 cycles (range 3–14) with an average of 145 (range 73–386) days of data analyzed per participant.

Results are presented for the data collected from 96 (71 complete and 25 partial) cycles. Menstrual cycles lasted a median of 28 days (IQR: 25–30 days) with menses lasting a median of 5 days. Four participants reported use of hormonal contraception (two oral contraceptives, one using a progesterone containing intrauterine device (IUD), and one using a progesterone-delivering implant).

Evaluation of CGM profiles is given in Table 1. The mean 24-h CGM glucose level was 161 ± 26 mg/dL during the menstrual phase, 165 ± 25 mg/dL during the luteal phase, and 159 ± 20 mg/dL during the rest of the cycle. Mean time in range (TIR) (70–180 mg/dL) was 69% ± 14% during the menstrual phase, 67% ± 13% during the luteal phase, and 69% ± 12% during the rest of the cycle. Analysis of intraparticipant glycemia revealed similar percentages of TIR between the luteal and menstrual phases (Fig. 1A) as well as similar percentages of time <70 mg/dL (Fig. 1B). Similar relationships were found when comparing all cycle phases (Supplement Fig. S1A, B).

FIG. 1.

(A) Intraparticipant CGM TIR 70–180 mg/dL during luteal^a versus menstrual^b phase. (B) Intraparticipant CGM time <70 mg/dL during luteal^a versus menstrual^b phases. ^aLuteal phases were estimated to occur during the 9 days before the onset of menses. ^bMenstrual phases were defined as contiguous days with reported vaginal bleeding, with a minimum 1 day and maximum 9 days observed during the study. CGM, continuous glucose monitoring; TIR, time in range.

Table 1.

Glycemic and Insulin Outcomes During the Menstrual Cycle (N = 16 Participants)

	Menstrual phase^a	Luteal phase^b	Rest of the cycle^c
CGM metrics
Hours of glucose readings	557 (414–676)	1059 (840–1250)	1334 (1065–1591)
Percent time in range 70–180 mg/dL	69% (14%)	67% (13%)	69% (12%)
Mean glucose—mg/dL	161 (26)	165 (25)	159 (20)
Percent time >180 mg/dL	30% (14%)	32% (14%)	29% (12%)
Percent time <70 mg/dL	1.3% (0.5%–2.3%)	1.3% (0.3%–1.8%)	1.4% (0.5%–2.0%)
Coefficient of variation—%	33% (30%–38%)	33% (30%–37%)	33% (30%–38%)
Insulin metrics (U/kg)
Total daily basal	0.28 (0.22–0.42)	0.28 (0.23–0.41)	0.28 (0.23–0.40)
Total daily bolus	0.29 (0.26–0.42)	0.31 (0.23–0.41)	0.29 (0.22–0.39)
Total daily insulin	0.65 (0.48–0.84)	0.62 (0.49–0.88)	0.60 (0.47–0.80)

Median (interquartile range) or mean (standard deviation).

Menstrual phases were defined as contiguous days with reported vaginal bleeding, with a minimum 1 day and maximum 9 days observed during the study.

Luteal phases were estimated to occur during the 9 days before the onset of menses.

Rest of cycle would include the follicular phase and ovulation if applicable.

CGM, continuous glucose monitoring.

Insulin delivery metrics throughout the cycle are given in Table 1. Median basal rates were 0.28 U/kg (IQR: 0.22–0.42) during the menstrual phase, 0.28 U/kg (0.23–0.41) during the luteal phase, and 0.28 U/kg (0.23–0.40) during the rest of the cycle. Median daily bolus doses were 0.29 U/kg (IQR: 0.26–0.42) during the menstrual phase, 0.31 U/kg (0.23–0.41) during the luteal phase, and 0.29 U/kg (0.22–0.39) during the rest of the cycle.

Further evaluation of changes in insulin delivery during specific time blocks during the 24-h period did not reveal any trends in total insulin delivery (Supplementary Fig. S2). Analysis of intraparticipant basal and bolus doses per kilogram delivery between luteal and menstrual phases showed similar delivery rates for both phases (Supplementary Fig. S3), with similar findings noted when comparing the other cycle phases (Supplementary Fig. 4A, B).

Discussion

Many prior publications and patient reports support that at least a subset of women with T1D experience changes in glucose levels and insulin requirements during different phases of the menstrual cycle.^5,7,8,10,11 Glucose and insulin metrics for our sample of 16 women using CLC were similar during different phases of the menstrual cycle. Evaluation of intraparticipant data during different cycle phases showed similar basal and bolus delivery and glycemic metrics throughout the cycle.

Though based on literature, there appears to be a subset of women who have variability in glycemic and insulin metrics across their cycle, this variability is expected to be smaller than the variability among participants. The number of participants studied in this analysis is larger than most other studies but would still not be expected to be robust enough to detect differences or changes in mean glycemic and insulin metrics among or within participants. It is possible that overall improvement in glycemic control may have blunted menstrual-associated changes in insulin sensitivity in the luteal phase leading to less robust changes in insulin delivery and glycemic variability.^2,7

Both Barata et al.¹⁰ and Brown et al.¹¹ reported HbA1c levels for the women they studied, but the relationship between glycemic control and changes in insulin resistance during these participants' cycles was not reported. Although the impact of glycemic control on changes in glucose control and insulin requirements around the menstrual cycle has not been clearly established, prior studies have reported that adolescents with higher HbA1c values had greater variability in the timing of their menstrual cycles.^15,16

The lack of intraparticipant changes in glycemia or insulin delivery was not expected, but more strategic delivery of insulin by the CLC controller as opposed to reactive changes in delivery in an open loop setting may have mitigated meaningful differences in glycemic outcomes at different times in the cycle.^15,16

The CLC system used in this study has been shown to improve TIR overall for people with T1D.¹² This secondary analysis found that this finding persisted when looking specifically at certain menstrual phases, and there was no phase in which the system universally performed poorly. The CLC system basal rate delivery can vary by 0 to ∼4 times the programmed basal rate. Changes required to mitigate the hormonal effects of the menstrual cycles, especially to basal rates, were likely much smaller than the range of changes allowed by the algorithm and thus may not have been pronounced enough for significant differences in insulin delivery or glycemic control to be detected.

Strengths of our study include a consistent prospective method for menstrual cycle reporting, and the uniqueness of our data set that reports both CGM and automated insulin delivery results. Limitations include that this is a secondary analysis that was not powered to find significant differences in glycemic or insulin metrics, a variable range of evaluated cycles among participants, and that there was overall excellent glycemic control in the studied population that could impact ability to generalize findings.

Collection of HbA1c levels was not performed at the time of enrollment into this substudy, since participants were not enrolled at a specified set point within the parent study. Participants used different methods of contraception, and no formal testing to evaluate ovulatory status was performed. Despite these limitations, the data we report support future studies to evaluate the use of CLC to address some women's struggles with menstrual cycle-related glycemic variability.

Emerging CLC systems should consider more detailed analyses in larger groups of women to ensure systems perform optimally for menstruating women and for those using different forms of contraception. Further studies formally designed to assess the impact of CLC in menstruating women and the impact of hormonal contraception and glycemic control are needed and could be better powered by enrolling women with self-reported difficulty in glucose control during particular phases of menstrual cycles and baseline glucose control stratification.

Overall, this study adds to an emerging literature of efforts to optimize glycemic control in menstruating women with T1D and provides reassurance that an available CLC system performs similarly across the menstrual cycle.

Footnotes

Authors' Contributions

C.J.L. wrote/edited the article. D.R. performed statistical analyses and wrote/edited the article. G.O.M., Y.C.K., L.M.L., J.E.P., J.W.L., and S.A.B. researched data, contributed to discussion, and reviewed/edited the article.

Acknowledgments

The iDCL Trial Study Group (site investigators noted): University of Virginia, Center for Diabetes Technology, Charlottesville, VA: Sue Brown (PI), Boris Kovatchev (Grant PI), Stacey Anderson (PI), Emma Emory, Mary Voelmle, Katie Conshafter, Kim Morris, Mary Oliveri, Linda Gondor-Fredrick, Harry Mitchell, Kayla Calvo, Christian Wakeman, Marc Breton; Joslin Diabetes Center, Harvard Medical School, Boston, MA: Lori Laffel (PI), Elvira Isganaitis (I), Louise Ambler-Osborn (I), Emily Flint, Kenny Kim, Lindsay Roethke; Sansum Diabetes Research Institute, Santa Barbara, CA: Jordan Pinsker (PI), Mei Mei Church (I), Camille Andre, Molly Piper; Division of Endocrinology, Diabetes, Icahn School of Medicine at Mount Sinai, New York City, NY: Carol Levy (PI), David Lam (I), Grenye O'Malley (I), Camilla Levister (I), Selassie Ogyaadu, Jessica Lovett; Division of Endocrinology, Diabetes, Metabolism, and Nutrition, Department of Internal Medicine, Mayo Clinic, Rochester MN: Yogish C. Kudva (PI), Vinaya Simha (I), Vikash Dadlani, Shelly McCrady-Spitzer, Corey Reid, Kanchan Kumari; Barbara Davis Center for Diabetes, University of Colorado, Anschutz Medical Campus, Aurora, CO: R. Paul Wadwa (PI), Greg Forlenza (I), G. Todd Alonso (I), Robert Slover (I), Emily Jost, Laurel Messer, Cari Berget, Lindsey Towers, Alex Rossick-Solis; Department of Pediatrics, Division of Pediatric Endocrinology and Diabetes, Stanford University School of Medicine: Bruce Buckingham (PI), Laya Ekhlaspour (I), Tali Jacobson, Marissa Town, Ideen Tabatabai, Jordan Keller, Evalina Salas; John A. Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge, MA: Francis Doyle III, Eyal Dassau; Jaeb Center for Health Research: John Lum, Roy Beck, Samantha Passman, Tiffany Campos, Dan Raghinaru, Craig Kollman, Carlos Murphy, Nandan Patibandla, Sarah Borgman; National Institute of Diabetes, Digestive, and Kidney Diseases (NIDDK): Guillermo Arreaza-Rubin (Project Scientist), Thomas Eggerman (Program Officer), Neal Green; iDCL Steering Committee Members: Boris Kovatchev, Sue Brown, Stacey Anderson, Marc Breton, Lori Laffel, Jordan Pinsker, Carol Levy, Yogish C. Kudva, R. Paul Wadwa, Bruce Buckingham, Francis Doyle III, Eric Renard, Claudio Cobelli, Yves Reznik, Guillermo Arreaza-Rubin, John Lum, Roy Beck; Central Laboratory—University of Minnesota Advanced Research and Diagnostic Laboratory: Robert Janicek, Deanna Gabrielson. Data Safety Monitoring Board (DSMB): Steven H. Belle (Chair), Jessica Castle; Jennifer Green, Laurent Legault, Steven M. Willi, Carol Wysham, Thomas Eggerman (DSMB Executive Secretary for NIDDK).

Author Disclosure Statement

C.J.L. reports receiving advisory board fees from Sanofi, and grant support, paid to her institution, from Dexcom, Tandem Diabetes Care, Insulet, Abbott Diabetes, Senseonics, and Lexicon Pharmaceuticals. G.O.M. receives research support from Tandem Diabetes, DexCom, and Abbot. D.R. has no financial disclosures. Y.C.K. received product support from Dexcom and Roche Diabetes and has consulted for Novo Nordisk. L.M.L. has received consulting fees from Johnson & Johnson, Sanofi, NovoNordisk, Roche, Dexcom, Insulet, Boehringer Ingelheim, ConvaTec, Medtronic, Lifescan, Laxmi, and Insulogic.

J.E.P. is currently an employee of Tandem Diabetes Care, Inc. The study presented in the article was performed as part of his academic appointment at Sansum Diabetes Research Institute and is independent of his employment with Tandem Diabetes Care. J.W.L. reports receiving consulting fees, paid to his institution, from Animas Corporation, Bigfoot Biomedical, Tandem Diabetes Care, and Eli Lilly. S.A.B. reports receiving grant support and supplies, provided to her institution from Tandem Diabetes Care, Insulet, and Tolerion, and supplies, provided to her institution, from Dexcom and Roche Diagnostics.

Funding Information

This study was funded by NIH/NIDDK grant UC4 108483. The University of Virginia Strategic Investment Fund Project No. 88 provided institutional and regulatory support. Tandem Diabetes Care provided the experimental closed-loop systems used in the trial, system-related supplies including the Dexcom CGM and Roche glucometer, and technical expertise. Tandem Diabetes Care was not involved in data analysis and was provided a copy of the article for review before publication.

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2

Supplementary Figure S3

Supplementary Figure S4

Supplementary Table S1

References

Cramer

: The influence of menstruation on carbohydrate tolerance in diabetes mellitus. Can Med Assoc J, 1942; 47:51–55.

Levy

, Widom

, Simonson

: Effect of the menstrual cycle on glucose metabolism and diabetes control in women with insulin-dependent diabetes mellitus. Semin Reprod Med, 1994; 12:110–118.

Lunt

, Brown

: Self-reported changes in capillary glucose and insulin requirements during the menstrual cycle. Diabet Med, 1996; 13:525–530.

DeFronzo

, Tobin

, Andres

: Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol, 1979; 237:E214–E223.

Toth

, Suthijumroon

, Crockford

, Ryan

: Insulin action does not change during the menstrual cycle in normal women. J Clin Endocrinol Metab, 1987; 64:74–80.

Yki-Jarvinen

, Koivisto

: Insulin sensitivity in newly diagnosed type 1 diabetics after ketoacidosis and after three months of insulin therapy. J Clin Endocrinol Metab, 1984; 59:371–378.

Widom

, Diamond

, Simonson

: Alterations in glucose metabolism during menstrual cycle in women with IDDM. Diabetes Care, 1992; 15:213–220.

Trout

, Rickels

, Schutta

, et al.: Menstrual cycle effects on insulin sensitivity in women with type 1 diabetes: a pilot study. Diabetes Technol Ther, 2007; 9:176–182.

Bergman

, Ider

, Bowden

, Cobelli

: Quantitative estimation of insulin sensitivity. Am J Physiol, 1979; 236:E667–E677.

10.

Barata

, Adan

, Netto

, Ramalho

: The effect of the menstrual cycle on glucose control in women with type 1 diabetes evaluated using a continuous glucose monitoring system. Diabetes Care, 2013; 36:e70.

11.

Brown

, Jiang

, McElwee-Malloy

, et al.: Fluctuations of hyperglycemia and insulin sensitivity are linked to menstrual cycle phases in women with T1D. J Diabetes Sci Technol, 2015; 9:1192–1199.

12.

Brown

, Kovatchev

, Raghinaru

, et al.: Six-month randomized, multicenter trial of closed-loop control in type 1 diabetes. N Engl J Med, 2019; 381:1707–1717.

13.

Brown

, Raghinaru

, Buckingham

, et al.: 101-LB: eighteen-month use of closed loop control (CLC): a randomized, controlled trial. Diabetes, 2020; 69(Suppl. 1):101-LB.

14.

Moglia

, Nguyen

, Chyjek

, et al.: Evaluation of Smartphone menstrual cycle tracking applications using an adapted APPLICATIONS scoring system. Obstet Gynecol, 2016; 127:1153–1160.

15.

Gaete

, Vivanco

, Eyzaguirre

, et al.: Menstrual cycle irregularities and their relationship with HbA1c and insulin dose in adolescents with type 1 diabetes mellitus. Fertil Steril, 2010; 94:1822–1826.

16.

Schroeder

, Hertweck

, Sanfilippo

, Foster

: Correlation between glycemic control and menstruation in diabetic adolescents. J Reprod Med, 2000; 45:1–5.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.08 MB

0.11 MB

0.10 MB

0.13 MB