Medpage Today reports on a study contradicting the rationale for delayed pushing, or “laboring down,” with an epidural. Analysis of the hospital records of 21,000 1st-time mothers concluded that delayed pushing resulted in longer 2nd-stage duration, longer active pushing phase, increased likelihood of cesarean and instrumental vaginal delivery, excessive postpartum bleeding, and more blood transfusions. By contrast, a recent Cochrane systematic review of randomized controlled trials (RCTs) (participants are allocated by chance to one form of treatment or the other) confirmed it. The review found that delayed pushing in women with epidurals results in longer 2nd-stage duration but shorter active pushing phase, that it tends to increase spontaneous vaginal birth, and that it has no effect on the incidence of excessive postpartum bleeding or any other adverse maternal outcome measured in the trials. One study found that 2 more babies per 100 would experience low blood pH at birth, but pooled data failed to find an increased risk of admission to neonatal intensive care. Which study has the right of it: the analysis of hospital records or the systematic review of RCTs?
To begin with, an RCT is the stronger design for this research question because RCTs set up a plan for how each group will be treated and then evaluate outcomes. In this case, women assigned to the “delayed” groups were told not to push unless they felt an uncontrollable urge, or the head was visible on the perineum (the block of tissue between the vagina and the anus), or until at least an hour (some trials prescribed longer delays) had passed after full dilation, and women in the “immediate” groups were directed to begin pushing at full dilation. A systematic review of RCTs is stronger yet because it pools data from multiple trials.
Investigators conducting the records analysis had to approximate a “delayed” versus “early” pushing group after the fact, which they did by defining “delayed pushing” as any labor in which pushing commenced 60 minutes or more after full dilation, “early pushing” as labors in which pushing began within 30 min after full dilation, and excluding women who began pushing between 31 and 59 minutes after full dilation.
This surrogate measure has several problems. The biggest is that women who develop a strong urge to push within the 1st hour or whose baby descends to the perineum within that time and are therefore instructed to push wouldn’t be counted in the “delayed” group. The review tells us this must happen fairly often. Average 2nd stage duration in trials in 1st-time mothers ranged from 1 hr 40 min to 3 hr 30 min, and it would likely take a while after beginning to push for 1st-time mothers with epidurals to birth their babies. Another difficulty is that a definition based on timing alone assumes that the reasons for delay don’t impact outcomes. The authors of the analysis acknowledge that this might not be the case, pointing out that pushing might be delayed in hopes of further descent or resolution of a malpositioned baby or because of maternal exhaustion. Similarly, while the RCTs evaluate outcomes from two groups distinguished by their differing treatment, the analysis chops a continuum of time arbitrarily into chunks and assumes an intentionality to those chunks. Furthermore, defining “early pushing” as within 30 minutes of full dilation isn’t the same as “immediate pushing,” as defined in the trials, which could also affect the validity of the analysis.
The analysis has yet another problem: all women in the RCTs have epidurals because the RCTs specifically evaluate a treatment for one of its harms: epidurals reduce spontaneous births because women don’t push as effectively. The analysis, however, includes women who don’t have an epidural.
For these reasons, then, the vote goes to the systematic review. For women with epidurals, waiting to push until either a strong urge develops or the head descends to the perineum appears to do little or no harm and is likely to do some good. Women without epidurals, it should be added, don’t need a policy. They should be aided in finding pushing positions other than reclining on the back and left to follow their bodies’ instincts as to when and how to bear down.