Utility and limitations of perimortem cesarean section: A nationwide survey in Japan

• Published in: The journal of obstetrics and gynaecology research 2019-02, Vol.45 (2), p.325-330
• Main Authors: Kobori, Shusaku, Toshimitsu, Masatake, Nagaoka, Shinichi, Yaegashi, Nobuo, Murotsuki, Jun
• Format: Article
• Language: English
• Subjects: Adult; Blood flow; cardiopulmonary arrest; Cardiopulmonary resuscitation; Cardiopulmonary Resuscitation - methods; Cardiopulmonary Resuscitation - statistics & numerical data; Cesarean section; Cesarean Section - methods; Cesarean Section - statistics & numerical data; Complications; CPR; Disseminated intravascular coagulation; Encephalopathy; Female; Heart Arrest - epidemiology; Heart Arrest - therapy; Humans; Hypoxia; Japan; nationwide survey; Neonates; Obstetrics; percutaneous cardiopulmonary support; perimortem cesarean section; Pregnancy; Pregnancy Complications, Cardiovascular - epidemiology; Pregnancy Complications, Cardiovascular - therapy; Pregnancy Outcome - epidemiology; Retrospective Studies
• ISSN: 1341-8076; 1447-0756
• DOI: 10.1111/jog.13819

Aim Perimortem cesarean section (PCS) is a procedure performed as part of cardiopulmonary resuscitation (CPR). This study aims to clarify maternal and neonatal prognosis and establish PCS's utility and limitations. Methods We sent structured questionnaires to obstetrics facilities regarding the cases of PCS performed in Japan between April 2010 and April 2015. Receiver operating characteristic (ROC) curve was used to determine the optimal cut‐off values of the time from cardiopulmonary arrest (CPA) to return of spontaneous circulation to predict the onset of disseminated intravascular coagulation (DIC). Results PCS was performed for 18 patients. Out of 18 patients, 12 were resuscitated. The women who were discharged without major sequelae (n = 6) were compared with the non‐discharged women who were dead or in persistent vegetative state (n = 12), and median interval time from arrest to PCS was significantly shorter in the former group (P = 0.002). Median interval time from CPA to PCS in the cases in which the neonates survived without major morbidities was significantly shorter than that in the cases of neonatal death and hypoxic encephalopathy (P = 0.01). DIC was observed in 8/9 (89%) patients whose resuscitation took more than 20 min from the diagnosis of CPA. Percutaneous cardiopulmonary support (PCPS) was introduced in 4/9 patients. However, more cases with uncontrolled bleeding, possibly caused by a sudden increase in blood flow and DIC after resuscitation, were observed in the PCPS group compared to the non‐PCPS group. Conclusion PCS can be an effective CPR procedure. However, if PCS is not performed within 20 min from CPA, starting PCPS before PCS is an option.