Does treatment modality affect measures of arterial stiffness in women with gestational diabetes?

OBJECTIVE: To investigate whether arterial stiffness (AS) differs between healthy women and women with gestational diabetes mellitus (GDM) managed by different treatment modalities. METHODS: This was a prospective longitudinal cohort study comparing AS in pregnancies complicated by GDM and low-risk controls. AS was assessed by recording aortic pulse-wave velocity (AoPWV), brachial augmentation index (BrAIx) and aortic augmentation index (AoAIx) using the Arteriograph® at four gestational-age windows: 24 + 0 to 27 + 6 weeks (W1); 28 + 0 to 31 + 6 weeks (W2); 32 + 0 to 35 + 6 weeks (W3) and ≥ 36 + 0 weeks (W4). Women with GDM were considered both as a single group and as subgroups stratified by treatment modality. Data were analyzed using a linear mixed model on each AS variable (log-transformed) with group, gestational-age window, maternal age, ethnicity, parity, body mass index, mean arterial pressure and heart rate as fixed effects and individual as a random effect. We compared the group means including relevant contrasts and adjusted the P-values using Bonferroni correction. RESULTS: The study population comprised 155 low-risk controls and 127 women with GDM, of whom 59 were treated with dietary intervention, 47 were treated with metformin only and 21 were treated with metformin + insulin. The two-way interaction term of study group and gestational age was significant for BrAIx and AoAIx (P < 0.001), but there was no evidence that mean AoPWV was different between the study groups (P = 0.729). Women in the control group demonstrated significantly lower BrAIx and AoAIx compared with the combined GDM group at W1-W3, but not at W4. The mean difference in log-transformed BrAIx was -0.37 (95% CI, -0.52 to -0.22), -0.23 (95% CI, -0.35 to -0.12) and -0.29 (95% CI, -0.40 to -0.18) at W1, W2 and W3, respectively. The mean difference in log-transformed AoAIx was -0.49 (95% CI, -0.69 to -0.30), -0.32 (95% CI, -0.47 to -0.18) and -0.38 (95% CI -0.52 to -0.24) at W1, W2 and W3, respectively. Similarly, women in the control group also demonstrated significantly lower BrAIx and AoAIx compared with each of the GDM treatment subgroups (diet, metformin only and metformin + insulin) at W1-W3. The increase in mean BrAIx and AoAIx seen between W2 and W3 in women with GDM treated with dietary management was attenuated in the metformin-only and metformin + insulin groups. However, the mean differences in BrAIx and AoAIx between these treatment groups were not statistically significant at any gestational-age window. CONCLUSIONS: Pregnancies complicated by GDM demonstrate significantly higher AS compared with low-risk pregnancies regardless of treatment modality. Our data provide the basis for further investigation into the association of metformin therapy with changes in AS and risk of placenta-mediated diseases.