MedsPregJC #1 – 20th November 2019

Welcome to the Medicines in Pregnancy Journal Club

@MedsPregnancyJC                     #MedsPregJC

 

Join us for our inaugural #MedsPregJC Twitter Journal club…..

Date: Nov 20th 2019 @ 9pm GMT.

Follow: @MedsPregnancyJC

Use the hashtag #MedsPregJC

Article citation: Dandjinou M, Sheehy O, Bérard A. Antidepressant use during pregnancy and the risk of gestational diabetes mellitus: a nested case–control study. BMJ open. 2019 Aug 1;9(9):e025908. https://bmjopen.bmj.com/content/9/9/e025908.abstract

TwitterJC host: Prof. Brian Cleary, Rotunda Hospital, Dublin, Ireland.

 

#MedsPregJC Summary

Aim: Determine the association between antidepressant classes, types and duration of use during pregnancy and the risk of gestational diabetes.

Study design: Nested case-control study.

Data source: Quebec Pregnancy Cohort, incorporating three Canadian databases which all pregnancies and children born in Quebec between 1998 and 2015.

Population selection: Cases of gestational diabetes (N=20905) identified after 20 weeks of pregnancy, randomly matched 1:10 with unaffected pregnancies (N=209050), matched for age and calendar year of delivery.

Inclusion criteria: All women in the Quebec Pregnancy Cohort (QPC) continuously covered for at least 6 months before and during pregnancy by the public prescription drug plan.

Exclusion criteria: Abortions and miscarriages, multiple pregnancies, women with type 1 and type 2 diabetes mellitus identified between 6 months before pregnancy and up to 20 weeks’ gestation, women with a history of GDM, women overweight or obesity (BMI >25kg/m²), cystic fibrosis.

Case selection: Gestational Diabetes Mellitus (GDM), defined as pregnant women with an ICD-9 or ICD-10 diagnosis code representing GDM or at least one filled prescription for an antidiabetic drug (Insulin, glyburide, metformin), both after week 20 of pregnancy.

Control selection: Random selection of 10 controls for each case among pregnancies that did not have a diagnosis of GDM at the index date.

Exposure:

1. Overall antidepressant use: (1) At least one prescription filled from first day of gestation to index date or a prescription filled before pregnancy and with a duration which included the first day of gestation (2) No antidepressant exposure (Reference).
2. Antidepressant exposure by class: (1) SSRI (2) SNRI (3) TCA (4) Other antidepressants (5) At least two antidepressant classes (6) No antidepressant exposure (Reference).
3. Antidepressant exposure by drug type: (1) Citalopram (2) Fluoxetine (3) Fluvoxamine (4) Paroxetine (5) Sertraline (6) Venlafaxine (7) Amitriptyline (8) Others (9) Combined antidepressants (1) No antidepressant exposure (Reference).
4. Antidepressant exposure by duration (number of days covered by prescription: (1) < 90 days (2) >90 days but <180 days (3) >180 days (4) No antidepressant exposure (Reference).

Confounders:

1. Sociodemographic and maternal characteristics: Maternal age, social assistance (Y/N), area of residence (Urban/rural)
2. Maternal co-morbidities in the 6 months prior to pregnancy: Depression/anxiety, chronic hypertension, asthma, thyroid disorders, cardiovascular disorders and polycystic ovarian syndrome based on physician-based diagnoses or filled prescriptions for related medications for chronic comorbidities.
3. Healthcare service utilisation and co-medication: Physician visits, hospitalisations or emergency department visits, number of other medication used other than antidepressants and medication related to chronic diseases (antipsychotics, benzodiazepines and corticosteroids) and psychiatrist visits within the last 6 months.

Analysis: Conditional logistic regression models were used to estimate crude and adjusted odds ratios.

Sensitivity analyses:

1. Restricted to women with depression/anxiety prior to pregnancy (diagnosis code for depression or anxiety or antidepressant prescription within 6 months before first day of gestation). Purpose: Take account for confounding by indication.
2. Restricted to diagnoses of GDM made within week 24 and 28 of pregnancy. Purpose: minimise non-differential bias.
3. Restricted to exposure between first day of gestation to 15 days before index date. Purpose: Ensure exposure preceded outcome.
4. Included and adjusted for women with a history of GDM. Purpose: account for potential confounding related to a history of GDM.
5. Generalised estimate equation (GEE) analysis. Purpose: Address intra-subject correlation.
6. Limited to women with overweight and obesity. Purpose: Women with overweight and obesity excluded from main analysis.

Results:

• Overall antidepressant use and risk of GDM: aOR 1.19 (95% CI 1.08-1.30).
• Risk of GDM varied according to AD class: SSRI aOR 1.07 (95% CI 0.96-1.20), SNRI aOR 1.27 (95% CI 1.08-1.48), TCA aOR 1.47 (95% CI 1.22-1.77), Other antidepressants aOR 1.06 (95% CI 0.83-1.36), combination of antidepressant aOR 1.38 (95% CI 1.15-1.67).
• Risk of GDM increased with longer duration of antidepressant use, specifically SNRI, tricyclic antidepressants, and combined use of two ADs.

Limitations: Potential for residual confounding by indication, analysis not adjusted for depression during pregnancy, smoking, weight gain, BMI, physical activity, income, education and ethnicity.

Author’s conclusions: The findings suggest AD’s, and specifically venlafaxine and amitriptyline were associated with increased risk of GDM.

Competing interests: Anick Bérard has served as a consultant for plaintiffs in litigations involving antidepressants and birth defects.