The recent trend in neonatal mode of delivery, in particular the increasing numbers of cesarean deliveries, has been the subject of considerable debate among health care providers.1 With few exceptions, the ideal mode of delivery for the term neonate is by spontaneous vaginal delivery. In cases in which this is not achieved, operative delivery—either by operative vaginal delivery or cesarean delivery—is required. Operative vaginal delivery (also called assisted or instrumental delivery2) describes delivery of the neonate with the assistance of either a forceps or a vacuum device. Although associated with a healthy neonatal outcome in the majority of cases, for many years concerns have been raised over the iatrogenic neonatal morbidity associated with instrumental delivery.3–6 However, the alternative approach for women who require expedited delivery—that is, a cesarean delivery at full dilatation—is not without risk.7 For this reason, both the American College of Obstetricians and Gynecologists and the Royal College of Obstetricians and Gynaecologists in the United Kingdom continue to endorse the use of instrumental delivery in appropriate cases.8,9
Rates of both operative vaginal and cesarean delivery are dependent on prevailing obstetric practice and thus vary significantly, both between different countries and over time. Recent figures from the U.S. government show that one in three neonates are delivered via cesarean.10 In contrast, operative vaginal delivery rates traditionally have been low and are decreasing.11 Despite global concern over the rapidly increasing rates of cesarean delivery, many clinicians continue to eschew the option of operative vaginal delivery.12 However, operative vaginal delivery is more widely practiced in Ireland and the United Kingdom, and relatively stable rates of 10–13% have been reported in recent years.9,13 We designed a cohort study to examine rates of peripartum neonatal mortality, neonatal encephalopathy, and neonatal death secondary to intracranial hemorrhage in term neonates over the course of a 10-year period in a single obstetric center with more than 1,000 operative vaginal deliveries per annum.
MATERIALS AND METHODS
This was a cohort study of 76,810 term singleton neonates 500 g or larger delivered at the National Maternity Hospital, Dublin, Ireland, over the course of the 10-year period from 2000 to 2009. The National Maternity Hospital is one of Europe’s busiest tertiary referral obstetric centers, with almost 10,000 deliveries per year. This was an analysis of data that are collated on an annual basis by a team of experienced clinicians in our institution and published in a comprehensive hospital clinical report. Detailed demographic data for all obstetric complications are collected annually. These reports, which include anonymized clinical details on all cases of perinatal death, fatal neonatal intracranial hemorrhage, and neonatal encephalopathy, were searched for the study period and data were extracted into a computerized database from which necessary additional details from individual patient records were retrieved.
Neonates of multiple gestations and neonates delivered preterm (less than 37 completed weeks of gestation) were excluded. Similarly, women who did not reach the second stage of labor—that is, elective and intrapartum cesarean deliveries performed before full dilatation---are excluded (Fig. 1). Cases involving two modes of operative delivery (ie, a double-instrumental delivery or a failed instrumental necessitating cesarean delivery) were not excluded and were classified under the initial modality; such “intention-to-treat” analysis is consistent with previous research on this issue.14 This is an analysis of anonymized hospital population data and therefore is deemed exempt from ethical approval from the National Maternity Hospital Research Ethics Committee. This study is reported according to the STROBE statement.15
Perinatal deaths were defined as “death of a fetus or neonate 500 g or larger during the pregnancy or in the first week of life.” Peripartum deaths were defined as “death of a fetus or neonate 500 g or larger intrapartum or in the first week of life,” that is, perinatal deaths excluding cases of antepartum stillbirths (which, by definition, cannot be associated with mode of delivery). Peripartum deaths secondary to lethal congenital anomalies were excluded, although it was not possible to exclude all cases of nonlethal fetal anomalies, because these could not be reliably extracted from our records. Deaths were classified as secondary to intracranial hemorrhage only when postmortem examination by a perinatal pathologist confirmed evidence of subdural, cerebral, intraventricular, or subarachnoid hemorrhage in a term neonate, which was considered the primary cause of death. Neonatal encephalopathy was defined as cases of term (37 weeks or more of gestation) neonates who “during the first week of life have either seizures alone or any two of the following lasting more than 24 hours: abnormal consciousness; difficulty maintaining respiration (of presumed central origin); difficulty feeding (of presumed central origin); and abnormal tone or reflexes.”
Women with uncomplicated pregnancies attending our institution are managed jointly with their primary health care provider. All women are offered a detailed mid-trimester fetal anomaly ultrasonogram. Screening for gestational diabetes mellitus (GDM) is performed at 28 weeks of gestation in women with historical risk factors, glycosuria, or fetal macrosomia or polyhydramnios. Women with a diagnosis of GDM in a previous pregnancy are treated as such in all future pregnancies. Induction of labor for postdates is routinely offered at 42 weeks of gestation in uncomplicated pregnancies. Suspected fetal macrosomia in the absence of GDM is not considered an indication for induction of labor. Elective cesarean delivery is considered in women with an estimated fetal weight of 4,500 g or more in GDM pregnancies (or 5,000 g or larger in the absence of GDM).
The decision for operative vaginal delivery in our unit and the choice of instrument are at the discretion of an experienced practitioner. Without exception, the decision for cesarean delivery is made by a consultant obstetrician. Operative vaginal delivery may be considered only at full dilatation and when the leading point of the skull is at or below the ischial spines.9 All operative vaginal deliveries are performed or directly supervised by a clinician with a postgraduate obstetric qualification and a minimum of 3 years of clinical obstetric practice. Manual rotation before application of an instrument is not commonly performed in our institution. In the early years of the present study, vacuum deliveries were performed almost exclusively with either a metal cup or a silastic cup. Since 2007, the majority of clinicians in our institution favor the Kiwi Omnicup as their instrument of choice for vacuum-assisted delivery. Nonrotational forceps deliveries are performed using either a Neville-Barnes forceps or a short Wrigley forceps, according to clinician preference. Rotational forceps deliveries, such as those using Kielland forceps, are not practiced at the National Maternity Hospital. Our institution follows a strict protocol for management of labor and delivery, which has not changed over the 10-year study period.16
The primary outcome of the present study was the rate of peripartum death by mode of delivery. Secondary outcomes were rates of neonatal encephalopathy and mortality related to neonatal intracranial hemorrhage by mode of delivery and the relationship between instrument choice and adverse outcomes. Cases of intracranial hemorrhage that did not result in neonatal encephalopathy or neonatal death could not be accurately identified from our database and are not included.
Statistical analysis was performed using Statsdirect 2.7.8. Categorical data were compared using the χ2 or Fisher exact test. When appropriate, odds ratios (ORs) and 95% confidence intervals (CIs) are included and are calculated relative to neonates delivered by both spontaneous vaginal delivery and cesarean delivery at full dilatation. The test for trends in continuous data was the Cuzick test for trend, which is a trend test for nonparametric data related to the Wilcoxon rank-sum test.17 Two-tailed P values were used throughout and the 5% level was considered significant.
Between January 1, 2000 and December 31, 2009, 83,889 neonates weighing 500 g or more delivered at the National Maternity Hospital. The total number of singleton neonates delivered at 37 weeks of gestation or more over the course of the 10-year study period was 76,810 (Fig. 1). Rates of cesarean delivery were 17.3% for the overall hospital population and 16.6% (n=12,764) for term singletons. Among cesarean deliveries, 36% were elective (prelabor), 60% were intrapartum before full dilatation, and 4% were intrapartum at full dilatation. The rate of operative vaginal delivery was 17% of women who delivered vaginally, or 14% of the overall hospital population (Fig. 1).
No significant trends in the rates of cesarean delivery (P=.08) or operative vaginal delivery (P=.08) were demonstrated over the course of the 10-year study period (Fig. 2). Among those undergoing operative vaginal delivery, approximately two-thirds (69%) underwent vacuum-assisted delivery and one-third (31%) underwent forceps-assisted delivery (Fig. 1). No significant trends in the contribution of forceps-assisted (P=.24) or vacuum-assisted (P=.77) deliveries to total instrumental deliveries were demonstrated (Fig. 2). In total, of women reaching the second stage of labor (ie, full cervical dilatation) in our institution (n=64,555), 82.3% underwent spontaneous vaginal delivery, 16.9% underwent operative vaginal delivery, and 0.8% underwent cesarean delivery at full dilatation. No significant trend in the rate of fully dilated women undergoing cesarean delivery was during the study period demonstrated (P=.41).
Among the women who reached full dilatation in our institution during the study period, the overall rate of peripartum fetal or neonatal mortality was 0.31 per 1,000 term singleton neonates (20/64,555). For neonates born by cesarean delivery in the second stage, the peripartum mortality rate was 1.96 per 1,000 (1/509). This was higher than seen in neonates born by spontaneous vaginal delivery (0.17/1,000; 9/53,113), but not significantly (P=.09), likely attributable to the small number of peripartum deaths in neonates delivered by second-stage cesarean (Table 1). The rates of peripartum mortality in total operative vaginal, vacuum-assisted, and forceps-assisted delivery were 0.8 (9/10,933), 1 (8/7,579), and 0.3 (1/3,354) per 1,000 neonates, respectively, which were not significantly different from that seen in neonates delivered by second-stage cesarean (Table 1).
The overall rate of neonatal encephalopathy, as defined, during the study period was 1.46 per 1,000 in women reaching full dilatation (94/64,555). In neonates delivered by second-stage cesarean, the neonatal encephalopathy rate was 3.9 per 1,000 (2/509). This was not different from the overall rate in neonates delivered by operative vaginal delivery (4.2/1,000; P>.99). The rates in neonates delivered by vacuum-assisted and forceps-assisted deliveries were 4.7 and 3.2 per 1,000, respectively, which were not significantly different from those born by second-stage cesarean (Table 2). For all three modes of operative delivery, the rate of neonatal encephalopathy was higher than the background rate in neonates delivered by spontaneous vaginal delivery, which was statistically significant for both vacuum-assisted (P<.001) and forceps-assisted (P=.007) deliveries (Table 2).
There were six cases of peripartum death secondary to neonatal intracranial hemorrhage in our institution between 2000 and 2009, for an overall incidence of 0.08 per 1,000 term neonates. The highest rate of intracranial hemorrhage--related mortality was seen in neonates born by vacuum-assisted delivery (0.4/1,000; 3/7,579). The equivalent rates in neonates born by forceps-assisted and spontaneous vaginal delivery were 0.3 (1/3,354) and 0.02 (1/53,113) per 1,000 neonates, respectively. Among 7,684 term neonates delivered by intrapartum cesarean in the first stage of labor, there was one death secondary to intracranial hemorrhage. There were no cases of neonatal death secondary to intracranial hemorrhage in neonates who underwent cesarean delivery at full dilatation.
Among neonates delivered by operative vaginal delivery (n=10,933), we examined the relationship between instrument (vacuum compared with forceps) and adverse peripartum outcomes. There was no difference in either overall peripartum mortality rates (OR 3.5, 95% CI 0.47–157; P=.29) or neonatal encephalopathy (OR 1.5, 95% CI 0.7–3.5; P=.38) in neonates born by vacuum compared with forceps delivery. The rate of intracranial hemorrhage--related mortality was 3–4 per 10,000 for both vacuum and forceps deliveries, with no significant difference between instruments (OR 1.3, 95% CI 0.1–69.7; P>.99), However, absolute numbers of adverse outcomes in both groups were small, and this study is likely underpowered for subanalyses by choice of instrument.
Rapidly increasing global rates of cesarean delivery have generated widespread debate. One of the contributing factors, particularly in the United States, has been the sharp decline in the proportion of women undergoing operative vaginal delivery.18 This was catalyzed by the 1998 Food and Drug Administration warning over the potential dangers associated with vacuum-assisted delivery.19 However, good-quality retrospective studies have not corroborated this finding. Towner et al5 found that rates of neonatal intracranial hemorrhage were similar for operative vaginal delivery and intrapartum cesarean delivery; they concluded that abnormal labor, rather than mode of delivery, was the likely culprit. Recently, Werner et al6 reported that composite neonatal neurologic morbidity was reduced after forceps deliveries compared with both vacuum and cesarean deliveries. Both of these studies were analyses of state-wide International Classification of Diseases, 9th Revision, Clinical Modification coded data and thus are limited by the weaknesses inherent in analyzing coded datasets.
In the present study, all neonates were delivered in a single obstetric center that utilizes a standardized labor management protocol.16 We found no difference in the rates of either peripartum mortality or neonatal encephalopathy in neonates delivered by vacuum-assisted, forceps-assisted, or second-stage cesarean delivery. Although the rate of neonatal encephalopathy was lowest in neonates delivered by forceps, consistent with the recent work by Werner et al,6 this was not significantly different from other operative deliveries. The finding that both options for instrumental vaginal delivery are associated with low (3–5/1,000) rates of neonatal encephalopathy lends further support to consensus statements from expert professional groups that highlight the role for instrumental delivery in selected women.8,9
Operative vaginal delivery is known to be associated with a range of potential complications.9 In addition to extracranial subgaleal bleeds,3,4 traumatic intracranial hemorrhage is the other most feared complication after operative vaginal delivery,20 with a reported incidence of 0.5–1.5 per 1,000 operative vaginal deliveries.9 In the present study, neonatal mortality secondary to term intracranial hemorrhage was examined. For both instruments, the rate of neonatal death from intracranial hemorrhage was 3–4 per 10,000, which was low. Although this rate appears higher than that for neonates born by spontaneous vaginal delivery, consistent with other studies5,21 the low number of cases across all modes of delivery precludes formal comparison for this rare adverse outcome. Furthermore, comparison with second-stage cesarean delivery is a more useful examination because this is the challenge faced by the clinician whose patient requires expedited delivery at full dilatation. Neonates born by intrapartum cesarean delivery also are vulnerable to intracranial trauma,5,7,22 with a neonatal death occurring from that in the present study.
Several authors have reported higher rates of neonatal injury after the use of sequential instruments in assisted delivery.5,9,23 Recently, Alexander and colleagues reported that morbidity was not higher in cases of failed instrumental delivery proceeding to cesarean compared with women proceeding directly to second-stage cesarean.24 We were unable to determine overall rates of neonatal complications in cases of sequential instrument use in our population, because such cases are not coded for in our hospital database.
We acknowledge several potential limitations to the present study. The retrospective design introduces the possibility of bias; however, these are data that are prospectively collated and published annually by a team of experienced clinicians, which should limit the potential for bias. Additionally, the large size of our dataset precludes the inclusion of accurate information on patient demographics, such as body mass index, medical comorbidities, chorioamnionitis, and others. Although a large study should limit this effect, the potential for confounding nonetheless remains. Second, for rare outcomes such as intracranial hemorrhage--related mortality, the total number of events in our population was low. Post hoc sample size estimates, assuming our rate of 4 per 10,000 vacuum deliveries, suggests that more than 30,000 instrumental deliveries would be required to demonstrate even a 10% difference in this outcome by mode of operative delivery. Similarly, although we demonstrated no statistically significant differences in peripartum mortality or neonatal encephalopathy between vacuum and forceps, both adverse outcomes were higher in neonates born by vacuum-assisted delivery, emphasizing the need for further study to determine the most appropriate choice of instrument.
Last, any study on the effects of mode of delivery on neonatal complication rates is limited by the inherent difficulty in separating indication for delivery from mode of delivery. Women undergoing operative delivery at full dilatation are more likely to have had difficult labors or nonreassuring fetal status, which will clearly contribute to the complication rates. This was confirmed by the higher rates of both peripartum mortality and encephalopathy after operative delivery compared with the background rate in spontaneous vaginal delivery. However, the comparison with second-stage cesarean delivery provides the clinician with more meaningful conclusions with which to inform day-to-day practice.
However, there are several strengths to our study that distinguish it from previous work on this issue. All our women delivered in a single tertiary obstetric center with a high volume of patients and a standardized labor management protocol. All operative vaginal deliveries are undertaken by experienced practitioners and, without exception, cesarean deliveries are sanctioned by a consultant obstetrician. The data included in the present study were extracted, not from a coded database, but from a prospectively collated annual clinical report, with individual patient records cross-checked as necessary. To our knowledge, this is one of the largest single-center studies to report neonatal outcomes by mode of delivery at term. It provides clinicians with important information regarding the effect of mode of delivery on neonatal outcomes. Furthermore, it demonstrates that operative vaginal delivery by an experienced obstetrician is a suitable choice in appropriate patients and is associated with similar rates of serious neonatal complications in second-stage cesarean delivery.
For women with term singleton pregnancies who require operative delivery at full dilatation, operative vaginal delivery and second-stage cesarean delivery are associated with similar rates of peripartum mortality and neonatal encephalopathy. The rate of neonatal death secondary to intracranial hemorrhage in neonates delivered by instrumental delivery is low (3–4/10,000). Operative vaginal delivery by an experienced practitioner is an appropriate choice in women who require expedited delivery.
The authors thank Fionnuala Byrne, Information Officer, National Maternity Hospital, for her help with data collection and retrieval.
1. Lumbiganon P, Laopaiboon M, Gulmezoglu AM, Souza JP, Taneepanichskul S, Ruyan P, et al.. Method of delivery and pregnancy outcomes in Asia: the WHO global survey on maternal and perinatal health 2007-08. Lancet 2010;375:490–9.
2. O'Mahony F, Hofmeyr GJ, Menon V. Choice of instruments for assisted vaginal delivery. The Cochrane Database of Systematic Reviews 2010, Issue 11. Art. No.:CD005455. DOI: 10.1002/14651858.CD005455.pub2.
3. Boo NY, Foong KW, Mahdy ZA, Yong SC, Jaafar R. Risk factors associated with subaponeurotic hemorrhage in full-term infants exposed to vacuum extraction. BJOG 2005;112:1516–21.
4. Doumouchtsis SK, Arulkumaran S. Head injuries after instrumental vaginal deliveries. Curr Opin Obstet Gynecol 2006;18:129–34.
5. Towner D, Castro MA, Eby-Wilkens E, Gilbert WM. Effect of mode of delivery in nulliparous women on neonatal intracranial injury. N Engl J Med 1999;341:1709–14.
6. Werner EF, Janevic TM, Illuzzi J, Funai EF, Savitz DA, Lipkind HS. Mode of delivery in nulliparous women and neonatal intracranial injury. Obstet Gynecol 2011;118:1239–46.
7. Murphy DJ, Liebling RE, Verity L, Swingler R, Patel R. Early maternal and neonatal morbidity associated with operative delivery in second stage of labour: a cohort study. Lancet 2001;358:1203–7.
8. American College of Obstetricians and Gynecologists. Operative vaginal delivery. ACOG Practice Bulletin 17. Washington (DC): ACOG; 2000.
9. Royal College of Obstetricians and Gynecologists. Green-top guideline No. 26. Operative vaginal delivery. London (United Kingdom): RCOG; 2011.
10. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Mathews TJ, Osterman MJ. Births: final data for 2008. Natl Vital Stat Rep 2010;59:1, 3–71.
11. Clark SL, Belfort MA, Hankins GD, Meyers JA, Houser FM. Variation in the rates of operative delivery in the United States. Am J Obstet Gynecol 2007;196:526.e1–5.
12. Jain NJ, Kruse LK, Demissie K, Khandelwal M. Impact of mode of delivery on neonatal complications: trends between 1997 and 2005. J Matern Fetal Neonatal Med 2009;22:491–500.
13. Begley C, Devane D, Clarke M, McCann C, Hughes P, Reilly M, et al.. Comparison of midwife-led and consultant-led care of healthy women at low risk of childbirth complications in the Republic of Ireland: a randomised trial. BMC Pregnancy Childbirth 2011;11:85.
14. MacDorman MF, Declercq E, Menacker F, Malloy MH. Neonatal mortality for primary cesarean and vaginal births to low-risk women: application of an “intention-to-treat” model. Birth 2008;35:3–8.
15. Von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008;61:344–9.
16. O’Driscoll K, Meagher D, Robson M. Active management of labour. 4th ed. Maryland Heights (MO): Mosby; 2003.
17. Cuzick J. A Wilcoxon-type test for trend. Stat Med 1985;4:87–90.
18. Oliphant SS, Jones KA, Wang L, Bunker CH, Lowder JL. Trends over time with commonly performed obstetric and gynecologic inpatient procedures. Obstet Gynecol 2010;116:926–31.
19. U.S. Food and Drug Administration. Public Health Advisory: need for caution when using vacuum assisted delivery devices. Rockville (MD): FDA Center for Devices and Radiological Health; 1998.
20. Gupta SN, Kechli AM, Kanamalla US. Intracranial hemorrhage in term newborns: management and outcomes. Pediatr Neurol 2009;40:1–12.
21. Demissie K, Rhoads GG, Smulian JC, Balasubramanian BA, Gandhi K, Joseph KS, Kramer M. Operative vaginal delivery and neonatal and infant adverse outcomes: population based retrospective analysis. BMJ 2004;329:24–9.
22. Dupuis O, Silveira R, Dupont C, Mottolese C, Kahn P, Dittmar A, et al.. Comparison of “instrument-associated” and “spontaneous” obstetric depressed skull fractures in a cohort of 68 neonates. Am J Obstet Gynecol 2005;192:165–70.
23. Murphy DJ, Macleod M, Bahl R, Strachan B. A cohort study of maternal and neonatal morbidity in relation to use of sequential instruments at operative vaginal delivery. Eur J Obstet Gynecol Reprod Biol 2011;56:41–5.
24. Alexander JM, Leveno KJ, Hauth JC, Landon MB, Gilbert S, Spong CY, et al.. Failedoperative vaginal delivery. Obstet Gynecol. 2009; 114:1017–22.