ObjectiveData and guidelines are lacking, so implantable cardioverter–defibrillators (ICDs) are often deactivated during labour to prevent inappropriate shocks. This study aimed to ascertain the safety of an activated ICD during labour.DesignAn observational study was performed.SettingDutch hospitals.Population or SampleA total of 41 childbirths were included of 26 patients who gave birth between February 2009 and November 2018 after receiving an ICD in our tertiary hospital. Five of these childbirths were attended by the research team between December 2018 and August 2020, during which the ICD remained active.MethodsGroups were made based on ICD status during labour. Patients who gave birth with an activated ICD at least once were stratified to the activated ICD group. Patients' files were checked and patients received a questionnaire about childbirth perceptions and treatment preferences. The differences in ordinal data resulting from the questionnaire were calculated using a chi‐square or Fisher's exact test.Main outcome measuresPrimary outcome was inappropriate ICD therapy and occurrence of ventricular arrhythmias requiring treatment.ResultsDuring the 41 childbirths, no inappropriate shocks or ventricular arrhythmias occurred during labour. All patients in the activated ICD group (n = 13) preferred this setting, while 8 of the 13 patients in the deactivated ICD group preferred activation (p = 0.002). Reasons included avoiding hemodynamic monitoring, magnet placement, or labour induction to facilitate technician availability.ConclusionsThis study shows no evidence that labour and birth in women with an activated ICD are unsafe, as there were no ventricular arrhythmias or inappropriate therapy. In addition, most patients prefer an activated ICD during labour.