- Background A rising proportion of maternal deaths are now attributable to cardiovascular disease. In pregnant women, postpartum cardiomyopathy (PPCM) or amniotic fluid embolism (AFE), and physiological changes of pregnancy such as increased blood volume and cardiac output, decreased systemic vascular resistance, and physiologic anemia may lead to cardiogenic shock (CS). The epidemiology and outcomes of CS during pregnancy are poorly characterized. We analyzed the National Inpatient Sample from 2002 to 2013 to determine the incidence, risk factors and outcomes of CS during pregnancy. Methods and results Of 52,973,186 antepartum or delivery, and 821,006 postpartum hospitalizations, 2044 were complicated by CS. CS occurred most often in the postpartum period (58.83%) as compared with delivery (23.47%) or antepartum (17.70%). The mortality rate in peripartum women with CS was 18.81% vs 0.2% without CS. Associated factors for CS during the antepartum or delivery period included PPCM, AFE, pulmonary thromboembolism, valvular disease, drug abuse, coagulopathy, fluid/electrolyte disorders, and peripheral vascular disorders. Factors associated with postpartum CS included PPCM, pulmonary thromboembolism, coagulopathy, fluid/electrolyte disorders, valvular disease, drug abuse, stroke, preeclampsia, and renal failure. Factors associated with death included cardiac arrest, renal failure, sepsis, younger age and delayed mechanical circulatory support. In women with CS, the use of mechanical circulatory support increased from 5.38% in 2002 to 14.2% in 2013. The mean time to mechanical support was longest during delivery (10.56 days, 95% CI: 2.55–18.56), followed by postpartum (4.95 days, 95% CI: 3.37–6.53) and antepartum (1.13 days, 95% CI: 0.21–2.06) hospitalizations. Overall, maternal survivors of CS received mechanical circulatory support sooner than those who died (4.36 days, 95%CI: 2.49–6.24 vs 12.46 days, 95%CI: 4.64–20.29, P < 0.001) . In women who received early mechanical circulatory support (defined as 6 days, the mean time to mechanical circulatory support), mortality was considerably less than in women who had increased times (>6 days) to mechanical circulatory support (18.08% vs 38.11%, OR = 0.29, 95% CI 0.1–0.86). Conclusions Although uncommon, CS is increasing during pregnancy, occurs most commonly in the postpartum period and is associated with dramatically higher mortality. The use of mechanical circulatory support in this critically ill population is increasing and early mechanical circulatory support may decrease mortality. Further work is needed to better characterize the causes of death from CS.