Is a prolonged second stage of labor too long?

Professional guidelines define the second stage of labor as ‘prolonged’ after 3 h for nulliparous women with an epidural and 2 h without, and after 2 h for parous women with an epidural and 1 h without1. Guidelines from the National Institute for Health and Care Excellence in the UK define prolonged second stage of labor (PSSL) with similar values, but take into consideration passive time at full cervical dilation and the time spent in active pushing2. In broader terms, these cut-offs approximate the 95th centile of the duration of second stage of labor3. As for other interventions4, the differential cut-offs for nulliparous and parous women, laboring with or without epidural anesthesia, became entrenched in practice based on clinical observation, without building an evidence base through properly performed studies. In recent years, the long-established guidelines for PSSL have come under renewed scrutiny5, 6. It has been suggested that these guidelines are arbitrary and do not lead to improved outcomes for women or their infants, but rather to an unnecessary increase in the rate of Cesarean delivery5, 7. Dietz and Campbell called for caution in the relentless efforts to reduce the rate of Cesarean delivery, citing the human cost of increased maternal and neonatal complications8. Conversely, Gimovsky and Berghella9 showed in a randomized trial that, when an extra hour was allocated to the standard guideline cut-off of duration of second stage of labor, the rate of Cesarean delivery was lower in the intervention arm. However, the actual duration of second stage of labor beyond the cut-off did not differ significantly between cases and controls, precluding cogent comparison of outcomes when extra time was allowed9. Many investigators argue that the present guidelines are not arbitrary and, rather, have shown that prolongation of the second stage of labor increases the risk of maternal and fetal complications10-21. For example, we showed that PSSL is associated with more perineal tears and that the risk for this increased significantly before the guideline-recommended cut-off for intervention was reached. PSSL and a large neonatal head circumference (HC; ≥ 35.5 cm) were both associated independently with increased risk of levator ani muscle (LAM) damage in primiparae22. PSSL exceeding 90 min doubled the risk of LAM damage compared with second stage duration of ≤ 60 min, while PSSL in excess of 160 min increased the risk for LAM damage 3.5-fold. When a neonate with a large HC was delivered following a PSSL, the risk for LAM trauma was increased more than 5-fold22. We further showed that LAM trauma and obstetric anal sphincter injury (OASIS) are strongly associated. Women with third- and fourth-degree anal sphincter tears had much higher rates of LAM trauma than did other gravidae23. We also demonstrated that a large HC and PSSL were more common in deliveries complicated by OASIS alone, LAM trauma alone and when the two were combined, than in women without pelvic floor trauma23. PSSL has been studied as an outcome in itself24-27 and as a factor implicated in the occurrence of other complications19, 28-31, such as operative delivery11, 14, 32, pelvic floor trauma19-21, 33-39, maternal hemorrhage40, 41, retained placenta42, low 5-min Apgar score10, 11, 31, low umbilical artery pH31 and admission to the neonatal intensive care unit (NICU)43, 44. PSSL affects 5–10% of all deliveries. Its associated complications are relatively rare overall, and can occur even when delivery is achieved before the guideline cut-off of second-stage duration is reached. A very large cohort is necessary to achieve statistically robust analyses to determine the effects of PSSL. With institutional review board approval and the help of our institution's Informatics Division, we established an electronic medical record-based database of births at our medical center, encompassing over 126 000 deliveries from 2003 to 2015, with an overall rate of Cesarean delivery (planned and unplanned, in primi- and multiparae) of 18%. We focused our investigation of PSSL on the subgroup of 19 111 primiparous women who received epidural anesthesia. Of these, 2506 (13.1%) exceeded the 3-h guideline cut-off of second-stage duration; as a general rule at our center, we are willing to manage extended labor expectantly where possible. These women had significantly increased rates of maternal hemorrhage (15.1% vs 10.2%, P < 0.001), 5-min Apgar score ≤ 7 (1.0% vs 0.6%, P = 0.05) and umbilical artery pH < 7.1 (2.4% vs 1.4%, P = 0.03), compared with those not exceeding 3 h. The rate of third- and fourth-degree tears to the anal sphincter (OASIS) was also increased (1.2% vs 0.9%, P = 0.058) and a non-significant increase in the rate of admission to the NICU (1.1% vs 0.8%, P = 0.11) was observed. Figure 1 shows the proportional frequency of these and other complications at 15-min strata for second-stage duration, from 60 min onwards. Complications are present at all strata but rise at 180 min. While there is no evidence to support shortening the second stage, it is also evident that extending the second stage is not without tangible risk to the mother and fetus. Our stratified complication rates show that, even at 120 min into the second stage of labor, the frequency of complications begins to increase above that of the reference stratum. Many fetuses are delivered for clinical indications, whether via vacuum extraction or Cesarean section, well before the 3-h guideline cut-off. The appearance of markers of acute fetal stress (umbilical artery pH ≤ 7.1 and 5-min Apgar score ≤ 7) at all time intervals in the second stage remind us to exercise caution before allowing the second stage to be further extended. Figure 2 shows the adjusted odds ratios of operative delivery, vacuum extraction and unplanned Cesarean delivery for each 15-min stratum of second-stage duration. The odds of an operative delivery began to increase noticeably before the 3-h guideline cut-off; the decision that intervention was necessary was made based on clinical findings in real time. Figure 3 compares the proportion of cases in each 15-min stratum of second-stage duration that were referred to Cesarean delivery for fetal distress and failure to progress as the primary reason. With increasing duration, the rate of fetal distress as the primary reason for Cesarean delivery decreased and the rate of failure to progress increased, with a crossover of proportions at 60–90 min. The diagnosis of failure to progress increased significantly before the 180-min guideline cut-off. PSSL is the final pathway for many complications in labor24-27; it brings to our attention that something is wrong which needs to be identified and acted on accordingly. It may be that the efforts of the mother and the tolerance of the fetus have reached a nadir: there is a problem with this delivery and it is time to intervene to avoid a graver outcome. If the mother and fetus are bearing up well when the second-stage duration cut-off is reached, there may be no need to intervene43, 45, 46. Guidelines provide limits that can be adapted to individual patient needs, but before we standardize an extended definition of PSSL, we must remember that serious complications often occur well before the present cut-off is reached. Although reducing the rate of Cesarean delivery is a worthy aim, as it may, in turn, lead to a reduction in maternal morbidity in future pregnancy, it cannot dictate individual patient management. Indeed, there is evidence that extension of the second-stage guideline cut-off will increase the risk of complications7, 12, 30, 31, 39, 47-49, and extension to avoid exceeding a quota of Cesarean deliveries7 may be unjustified8, 49.