Patient Safety in Obstetrics: More Evidence, Less Emotion
In this edition of Obstetrics & Gynecology (see pages 262 and 270), two articles address different aspects of obstetric thrombosis risk.1,2 Robison and colleagues describe the implementation of a risk-based enoxaparin administration protocol at their institution, comparing rates of use before the protocol and after.1 Not surprisingly, the rate of use of enoxaparin increased after the protocol (with an electronic order set) was implemented, and adherence to the protocol was high. There were no differences in thrombosis rates between the two study periods, but the study was underpowered to assess clinically important differences. Mardy and colleagues used the Perspective database to examine trends and variation in thromboprophylaxis in antepartum and postpartum patients.2 They found that the rate of thromboprophylaxis has increased in the United States over the past 10 years but that there is significant variation in its use at both the hospital and regional levels.
Although both of these studies are interesting and valuable additions to the literature, I believe that there is a much more central question that we need to address—what is the optimal strategy for thromboprophylaxis for our post–cesarean delivery patients? Recently, the National Partnership for Maternal Safety (a group that has representation from many national obstetrics and gynecology organizations) published a consensus bundle on thrombosis.3 The rationale for the need for the bundle came from two pieces of information. First, longitudinal data from the National Inpatient Sample suggests that rates of thrombosis during pregnancy have increased over the past 10 years despite increased use of mechanical thromboprophylaxis. Secondly, the Royal College of Obstetricians and Gynaecologists (RCOG) developed aggressive guidelines for pharmacologic thromboprophylaxis in 2004, and, since that time, maternal death from obstetric thromboses has decreased in the United Kingdom. I personally do not find that such analyses of trends, which are among the weakest of epidemiologic study designs to assess causality, make for a compelling case for change. Additionally, professional organizations can reasonably disagree—just because RCOG has recommendations for expanded pharmacologic thromboprophylaxis does not mean that we need to follow suit. Our countries differ in many ways, such as in populations and health care systems. What is effective in one country may not be effective in another. There are many examples of this being historically true; one example is the active management of labor. That care package was extremely effective in the United Kingdom, but the results could not be replicated in the United States.