Many Pregnancy-Related Deaths May Be Preventable

Approximately four out of every 10 pregnancy-related deaths in North Carolina are potentially preventable, new study findings suggest.

The fact that several other developed countries have maternal mortality ratios lower than that of the US, and that some racial and ethnic groups have higher risks of pregnancy-related death, led lead investigator Dr. Cynthia J. Berg and her colleagues to theorize that maternal mortality rates could be lower than they are.

The North Carolina State Center for Health Statistics combed through four computerized databases to identify 108 pregnancy-related deaths that occurred between 1995 and 1999. Dr. Berg and her team evaluated these cases and report their findings in the December issue of Obstetrics and Gynecology.

Based on the causes of death, Dr. Berg's group found that 41 cases could have been prevented.

Their analysis showed that 90% of deaths due to hemorrhage and chronic diseases could have been prevented, as could a substantial proportion of those caused by cardiomyopathy, pregnancy-induced hypertension, infection, cardiovascular condition and choriocarcinoma.

Only deaths due to cerebrovascular accident, amniotic fluid emboli, or microangiopathic hemolytic syndromes were considered unavoidable because of their catastrophic nature and lack of uniformly effective therapy.

The authors attribute 22 cases of preventable death to poor quality medical care and four cases to "nonoptimal organization of the health care system," such as failure to plan for follow-up or transfer of the patient.

Lack of preconception care, during which patients should have been advised about the risks associated with serious medical conditions, was responsible for eight deaths. Patient actions, such as failure to adhere to medical advice or to seek care in a timely fashion, were involved in 14 deaths.

"In-depth review of pregnancy-related deaths can help determine strategies needed to continue making pregnancy safer," Dr. Berg and her associates conclude.

Obstet Gynecol 2005;106:1228-1234.