Commentary: How to respond to the trend of increasing cesarean delivery

June 1, 2010

In 2007, 31.8% of deliveries in the United States were accomplished by cesarean delivery, with more than 1,370,000 women undergoing the procedure. This marks a 2% increase over 2006 and a more than 50% increase since 1996. This figure is not accounted for by repeat procedures only.

Key Points

In 2007, 31.8% of deliveries in the United States were accomplished by cesarean delivery (CD), with more than 1,370,000 women undergoing the procedure.1 This represents a 2% increase over 2006 and a more than 50% increase since 1996. This increase in CD is not accounted for by repeat procedures only. In 2004, one-fifth of all deliveries were primary cesareans, a 41% increase in this category since 1996.2 International rates also increased 62% between the epochs of 1993 to 1997 and 2003 to 2007, ranging from 6% to 211% over baseline (Figure).1,3-6

In which of these categories could a safe vaginal delivery have been achieved? Are we assiduously applying evidence-based standards to assess progress in labor? Have we mastered fetal heart rate interpretation (despite its limitations), and are we using other biophysical parameters (scalp or acoustic stimulation) to reduce false-positive diagnoses of fetal intolerance?10 Could better surveillance of fetal lie combined with more consistent application of external cephalic version (ECV) reduce breech presentation?11 Should we reconsider vaginal breech delivery by applying a standardized protocol and by training or retraining obstetricians to maximize fetal safety?12 Finally, since we cannot accurately predict macrosomia, should our current breakpoints of estimated fetal weight that trigger primary CD be reconsidered, or should patients with suspected macrosomia labor with lower thresholds to define arrest?

Listing these questions does not imply that easy solutions exist. However, not exploring these issues is a tacit acceptance of excess maternal risk that is potentially avoidable.

What follows are abbreviated reviews of arrest of labor, arrest of descent, fetal intolerance, management of breech, and suspected macrosomia to identify relevant clinical questions in each category. Reevaluating TOL is also discussed as a secondary strategy, since avoiding primary CD will not be possible in many instances. Finally, CD on patient request is also addressed.

Arrest of dilation

In the Maternal-Fetal Medicine Unit Network's Cesarean Delivery Registry, arrest of labor accounted for 59% of CDs during the first stage of labor and 84% performed during the second stage.13 The current American College of Obstetricians and Gynecologists (ACOG) Practice Bulletin on Dystocia and Augmentation of Labor describes active-phase arrest as 2 hours of adequate contractions without cervical change.14 However, patients who are permitted additional time in active labor can still deliver vaginally without significantly increasing maternal or fetal morbidity.15 Adhering to rigid criteria has the potential to cause overdiagnosis of arrest disorders.16 Effective dilation culminating in vaginal delivery can occur more slowly than previously described and can even include 2-hour periods without cervical change.17

Women who receive oxytocin augmentation progress at a slower rate compared with women who spontaneously labor.18 When this latter group had labor arrest for at least 2 hours, providing them with an additional 4 hours of augmentation was found to be safe and effective, with 92% of women ultimately delivering vaginally and without adverse maternal or fetal outcomes.15

Among women with no progress after even 4 hours of oxytocin, 88% of parous women and 56% of nulliparous women ultimately delivered vaginally, although they did experience a higher rate of infection compared with their quicker-to-deliver counterparts.15 For women undergoing oxytocin induction at term, allowing a latent phase of at least 18 hours was found to be safe and effective.19

The possibility that physicians perform abdominal delivery for "arrest" sooner than is appropriate has been reported in a study of Canadian providers.20 Among 239 laboring nulliparas delivered by 10 physicians, relevant guidelines were followed less than half the time in spontaneous labors ending with CD.

A study of labor in community hospitals in Los Angeles County and Iowa found that one-third of 231 CDs performed for lack of progress did not meet the definition of active-phase arrest according to ACOG criteria.21 It would appear, then, that more consistent application of current labor criteria has the potential to reduce the rate of primary CD.

Arrest of descent

Another opportunity to reduce primary CD is the appropriate use of operative vaginal delivery (forceps and vacuum extraction). A randomized trial of operative vaginal delivery versus CD has not been conducted. Nonetheless, operative vaginal delivery can provide an alternative to CD that decreases maternal morbidity and is equally safe for the fetus in properly selected cases. Since 1996, the rate of operative vaginal delivery has been reduced by half.3 Safe use of these devices requires adequate training and, most important, sound clinical judgment.

Successful rotation from occiput posterior to occiput anterior position is associated with a reduction in the rates of CD and perineal injury.22 However, rotation and operative vaginal delivery have been largely abandoned. Yet there is no evidence that either rotation or operative vaginal delivery per se, rather than improper case selection, is to blame for fetal or maternal injury. We would suggest that more thoughtful management of second-stage arrest attributed to fetal malposition could prevent a subset of primary CDs.