Effects of Chewing Gum on Glycaemic Control in Women With Gestational Diabetes
Effects of Chewing Gum on Glycaemic Control in Women With Gestational Diabetes
This study aims to assess the impact of enhanced chewing on glycaemic control in females with newly diagnosed GDM. It is hypothesised, that a fixed amount of gum chewed for 20 minutes before starting each meal could improve hyperglycaemia. The impact of chewing on postprandial capillary blood glucose (measured at one hour after breakfast, lunch and dinner) is determined as the primary outcome of this study.
Differences in fasting glucose and longitudinal changes over the study period should be additionally examined.
Study design and participants This study is designed as an open-label, mono-centre randomized controlled trial with two parallel groups including a total of 74 female patients with recent diagnosis of GDM. Diagnosis of GDM is made in accordance with the IADPSG criteria between 24+0 and 27+6 weeks of gestation [IADPSG 2010]. All pregnant females (aged
between 18 and 45 years) will be recruited consecutively among women visiting our pregnancy outpatient department (Department of Obstetrics and Gynecology, Division of Obstetrics and fetomaternal Medicine, Medical University of Vienna).
Exclusion criteria
Treatment and Interventions Eligible patients are randomized to receive either routine care (control group) or routine care in addition to a chewing gum intervention for five days (beginning with the second study day, i.e. the day after baseline examination): sugar-free, fruit or mint flavoured gum chewed three times daily for 20 minutes before each meal. Routine care includes standard dietary and lifestyle advice for 30 minutes following our local recommendations as well as an advice on capillary blood glucose measurement
(fasting as well as 1h after starting each meal). Capillary blood glucose profiles are re- evaluated seven days after starting the blood glucose monitoring.
Study visits will be scheduled at screening (visit 1, where eligible patients were randomized) as well as seven days later (visit 2).
Randomisation Participants will be randomized to either treatment (chewing gum) or control group (routine care) in a 1:1 ratio. The minimisation method [Pocock 1975] will be used to minimize the imbalance between the groups according to the preconceptional overweight/obesity status with three strata: i. normal weight (i.e. BMI below 25 kg/m2); ii. overweight (BMI 26 - 30 kg/m2); iii. obesity (BMI and above 30 kg/m2).
Medical history and baseline examination A broad risk evaluation will be performed in participating females at the initial contact (between 24+0 and 32+0 weeks of gestation) including: evaluation of maternal age, parity, history of GDM, detailed family history, ethnicity, preconceptional diseases, use of ovulation drugs, obstetric history. Moreover, an evaluation of preconceptional weight and BMI as well as measurement of blood pressure will be performed.
Assessment of dietary patterns Dietary patterns will be assessed at baseline (visit 1) by using a published and validated Food-Frequency-Questionnaire (FFQ) proposed by the German Robert Koch Institute [Haftenberg 2010]. It was also previously used for the German DEGS project (www.degs-studie.de). Information from the FFQ will be analyzed quantitatively or summarized by eating scores proposed in the literature (such as the Healthy Eating Index 2010 or Alternate Healthy Eating Index 2010) reflecting diet quality based on actual guidelines [Guenther 2013, Chiuve 2012]. In addition all patients will be advised to conduct a nutritional protocol for seven days. D. End Points
The three primary end points of this study are the averages (arithmetic mean values) of five days 1h postprandial capillary blood glucose measurements: after breakfast (primary end point 1), lunch (primary end point 2) and dinner (primary end point 3).
Secondary end points are fasting glucose concentrations and longitudinal changes (i.e.
assessment of group by time interactions in fasting and postprandial glucose trajectories) as well as changes in dietary patterns.
E. Statistical Analysis
Sample size With a sample size of n=35 subjects per group we are able to detect a mean difference of 8 mg/dl in postprandial glucose levels with a power of 81% and a type 1 error of α=0.016 for a two-sided unpaired student's t-test, providing a standard deviation of 10 mg/dl in accordance with [Landon 2009]. In order to achieve a 95% coverage probability, the Bonferroni correction will be used to adjust for multiple testing (three primary outcomes). Considering a drop-out rate of 5% n=74 cases are required for this study. A sample size review and adaptation is planned after 50% of the subjects have been investigated.
Analysis Plan Categorical variables will be summarized by counts and percentages; continuous variables data are summarized by means and standard deviations (SD) or by median and interquartile range in the case of strong deviations from the normal distribution.
Statistical comparison of continuous parameters (e.g. postprandial or fasting glucose concentrations) will be performed by the two-sample t-test or nonparametric
approaches (e.g. the Brunner-Munzel test), respectively. An adjustment for demographic variables (such as age) will be performed in case of significant group differences using analysis of covariance (ANCOVA) or by the proportional odds model, if the normality assumption is violated. Linear mixed effects models will be used to assess longitudinal changes of parameters of interest (i.e. fasting and postprandial glucose). Interaction terms are included in the longitudinal models to assess group differences and variable transformations (e.g. logarithmic or square root transformation) are used if necessary. A two-sided p-value ≤0.05 is considered statistically significant. Using the Bonferroni correction for multiple statistical testing (three hypotheses) this results in a 98.33% confidence interval to achieve a 95% simultaneous coverage probability. All analyses will be performed by using the statistic software R and contributing packages [R Core Team 2016].
Inclusion Criteria:
-
Exclusion Criteria: