Severe neonatal birth injury: Observational study of associations with operative, cesarean, and spontaneous vaginal delivery

To determine the association of successful and unsuccessful operative vaginal delivery attempts with risk of severe neonatal birth injury.


INTRODUCTION
Neonatal birth injuries are prevalent, but risks associated with mode of delivery are poorly characterized. 1,2Severe birth injury occurs in every five live births per 1000. 2 Data from large cohorts of pregnant women suggest that the prevalence of neonatal birth injury is increasing with time, including severe and mild injuries. 3Part of the increase is thought to relate to the growing use of operative vaginal delivery, which is encouraged to prevent cesarean section. 3Yet, the potential to increase the proportion of neonates with severe injuries is unclear.][10][11] These studies suggest that infants born by operative vaginal delivery have two to eight times the risk of any severe injury compared with cesarean delivery. 9,11Forceps tends to be associated with severe birth injuries, although vacuum appears to increase the risk of subarachnoid hemorrhage. 8Less is known about unsuccessful operative vaginal attempts that lead to emergency cesareans.In one analysis, infants delivered by cesarean after an unsuccessful operative attempt had nine times the risk of intracranial hemorrhage compared with spontaneous delivery. 10However, the association with other types of severe injury is unknown.We studied the extent to which successful and unsuccessful operative vaginal attempts were associated with the risk of severe birth injury compared with spontaneous vaginal delivery.

Study population
We conducted a population-based observational study of 1 080 503 infants born in hospitals of Quebec, Canada between 2006 and 2019.To carry out the study, we retrieved patient data from the Maintenance and Use of Data for the Study of Hospital Clientele registry. 12The registry compiles discharge abstracts for 98% of deliveries in Quebec and contains up to 41 diagnostic and 35 procedure codes for mothers and newborns.Diagnoses are coded using the 10th revision of the International Classification of Diseases, and procedures using the Canadian Classification of Health Interventions.Infant charts are linked with maternal charts, allowing us to identify the mode of delivery for each newborn.We did not include stillborn infants in this study as birth injury data are not collected for this population.

Mode of delivery
The main exposure measure was mode of delivery, including operative vaginal delivery with forceps or vacuum, cesarean delivery with or without an operative vaginal attempt, and spontaneous vaginal birth.We classified the mode of delivery into the following categories: forceps only, vacuum only, use of both forceps and vacuum, elective cesarean, emergency cesarean with no operative vaginal attempt, and emergency cesarean after an unsuccessful operative vaginal attempt.The reference group included infants born by spontaneous vaginal delivery.To identify the mode of delivery, we used procedure codes from maternal charts (Table S1, Supporting Information).Procedure codes allowed us to distinguish elective from emergency cesarean deliveries beginning in 2009.

Neonatal birth injury
The main outcome measure was severe mechanical birth injury, defined as any severe injury due to physical trauma from the birthing process.We included intracranial hemorrhage (subdural, cerebral, other), brain and spinal damage, Erb's paralysis and other brachial plexus injuries, epicranial subaponeurotic hemorrhage, skull and long bone fractures, and liver, spleen, and other body injuries. 3We identified injuries following diagnostic codes in the International Classification of Diseases-10 and did not include injuries that were not due to external forces (Table S1).We analyzed severe injuries as a composite indicator, as well as individually.
As a secondary outcome, we included an indicator for any mechanical birth injury without restricting to severe injuries only.This indicator accounted for severe injuries

Statistical analysis
We calculated the prevalence of birth injury per 1000 infants.We estimated risk ratios (RR) for the association of delivery mode with specific types of injury using logbinomial regression models.Models were adjusted for maternal age, parity, infant sex, preterm birth, multiple birth, socioeconomic disadvantage, place of residence,  and time period.We used generalized estimating equations to account for neonates with the same mother. 14n sensitivity analyses, we examined the association between mode of delivery and severe neonatal injury according to birth weight (low <2500, normal 2500-3999, high ≥4000 g).We also examined trends by time period owing to potential improvements in clinical practice over time.
We carried out the analysis in SAS v9.4 (SAS Institute Inc., Cary, NC) and used 95% confidence intervals (CI) to assess statistical significance.The dataset was anonymized.We received an ethics waiver from the institutional review board of our institution and followed the principles of the Declaration of Helsinki.

RESULTS
The cohort included 1 080 503 neonates between 2006 and 2019, including 101 312 (9.4%) born by operative vaginal delivery and 265 390 (24.6%) born by cesarean (Table 1).Operative vaginal delivery was more common for women <25 years and women who delivered at term.Cesarean delivery was more frequent for women ≥35 years and women with preterm delivery.Both operative vaginal and cesarean delivery were more common in nulliparous women.
Operative vaginal delivery was associated with risk of severe birth injury (Table 2).Compared with spontaneous delivery, forceps (RR 3.35, 95% CI 3.07-3.66)and vacuum (RR 2.98, 95% CI 2.80-3.16)were each associated with an elevated risk of severe injury.Risk of severe birth injury was significantly greater when forceps and vacuum were used sequentially (RR 8.69, 95% CI 7.70-9.82).Failed operative vaginal attempts that resulted in an emergency cesarean were associated with 1.55 times the risk of severe birth injury (95% CI 1.36-1.76).However, cesarean deliveries in which there was no operative attempt were not associated with severe birth injury.Associations were similar when we analyzed mild and severe injuries together, although vacuum delivery was associated with a greater risk of any injury (RR 5.65, 95% CI 5.53-5.77)than forceps (RR 3.89, 95% CI 3.75-4.03).
Operative vaginal delivery was the most common mode of birth among infants with intracranial hemorrhage (54.7%) and epicranial subaponeurotic hemorrhage (65.0%) (Figure 1).Operative vaginal delivery accounted for a third of brain and spinal damage (39.6%) and Erb's and other brachial plexus injuries (30.2%).Cesarean Half of all severe injuries occurred after a spontaneous vaginal delivery (45.6%), followed by 40.6% after operative vaginal and 13.8% after cesarean delivery.
Associations between cesarean delivery and severe birth injury were not as strong (Table 4).Cesarean delivery was associated with a slight risk of any severe injury, compared with spontaneous delivery (RR 1.11, 95% CI 1.04-1.18).
The association appeared to be driven by emergency cesarean after an unsuccessful operative vaginal attempt, which was associated with intracranial hemorrhage (RR 2.72, 95% CI 1.37-5.39),epicranial subaponeurotic hemorrhage (RR 3.20, 95% CI 2.33-4.38),and liver, spleen, and other body injuries (RR 2.69, 95% CI 2.11-3.42).Elective cesarean and emergency cesarean delivery without an operative vaginal attempt were not associated with severe injury, except liver, spleen, and other body injuries.
In sensitivity analyses of birth weight, operative vaginal delivery was associated with severe injury among infants with low (RR 2.80, 95% CI 1.93-4.06),normal (RR 3.37, 95% CI 3.17-3.57),and high weight (RR 2.91, 95% CI 2.58-3.28),compared with spontaneous delivery (Table S2).Emergency cesarean following an operative vaginal attempt was associated with a greater risk of severe injury among infants with normal birth weight (RR 1.40, 95% CI 1.20-1.64),but not other infants.There was no evidence of a change in the association with severe injury over time.

DISCUSSION
This population-based study of 1.1 million infants suggests that successful and unsuccessful operative vaginal attempts are both associated with a greater risk of severe birth injury compared with spontaneous delivery.Forceps delivery is strongly associated with intracranial hemorrhage, brain and spinal damage, and severe liver, spleen, and other body injuries.Vacuum delivery is in contrast associated with epicranial subaponeurotic hemorrhage and skull fractures.Operative vaginal attempts that result in an emergency cesarean are associated with the risk of intracranial hemorrhage, epicranial subaponeurotic hemorrhage, and liver, spleen, and other body injuries.Overall, the results suggest that operative vaginal attempts are a strong risk factor for severe birth injury, even when the attempt is unsuccessful.
Prior studies have rarely contrasted operative vaginal delivery against spontaneous birth, focusing instead on the comparison of operative delivery with cesarean delivery. 9,11There are a few exceptions. 8,10,15An analysis of 119 432 infants from Nova Scotia found that forceps was associated with four times and vacuum three times the risk of severe birth injury compared with spontaneous vaginal delivery, but specific injuries were not examined. 15In studies that did examine different types of injury, forceps and vacuum were associated with two to five times the risk of intracranial hemorrhage and brachial plexus injuries compared with spontaneous delivery. 8,10These studies were, however, conducted decades ago when obstetric practice may have differed. 8,10In our data, forceps and vacuum were associated with more than 10 times the risk of intracranial hemorrhage compared with spontaneous delivery, an association that is considerably greater than those measured in past studies. 8,10isk of severe injury was greater when both forceps and vacuum were used.Sequential use of forceps and vacuum was associated with more than 59 times the risk of intracranial hemorrhage and epicranial subaponeurotic hemorrhage compared with spontaneous delivery.In other studies, combined use of forceps and vacuum was associated with three to eight times the risk of intracranial hemorrhage and brachial plexus injuries compared with spontaneous delivery. 10,16These studies relied on less comprehensive definitions of severe birth injury or were restricted to nulliparous women, 10,16 which may explain why the associations were not as strong as ours.Our findings suggest that risks associated with joint use of forceps and vacuum may have been significantly underestimated in past studies.
It is generally thought that cesarean delivery protects against birth injury. 10,15,17Previous studies indicate that elective cesarean is associated with an 80% lower risk of severe injury compared with spontaneous delivery. 17evertheless, associations may not be as protective for emergency cesareans.A study of 583 340 neonates from California found that emergency cesareans that were not preceded by an operative vaginal attempt were associated with two times the risk of intracranial hemorrhage compared with spontaneous delivery. 10,15Emergency cesareans after an unsuccessful operative attempt were associated with more than four times the risk. 10,15In our data, cesareans that were not preceded by an operative attempt were generally not associated with severe injury.Rather, emergency cesarean after an unsuccessful operative attempt was associated with most types of severe injury.Overall, however, the risk of severe injury was not as great compared with combined use of forceps and vacuum.
6][7] The American College of Obstetricians and Gynecologists and Society of Obstetricians and Gynecologists of Canada do not recommend using forceps and vacuum together, unless one instrument cannot be applied optimally or a difficult cesarean is expected. 5,6UK guidelines stress the risk of severe injury with use of two instruments, but discourage cesareans nevertheless. 7However, these guidelines were developed from past data that may have underestimated the risks associated with forceps and vacuum.Our data indicate that sequential use of instruments may be associated with more than 15 times the risk of severe injury.Thus, guidelines for use of cesarean section after an unsuccessful operative vaginal attempt may benefit from re-evaluation.
Related risk factors for severe birth injury may also need to be considered.Breech presentation, malposition, and prolonged labor are all associated with an increased risk of severe injury. 11,15,18,19These factors may influence the choice of delivery mode. 11,15,18,19A Swedish study of 27 357 breech and 837 494 cephalic deliveries found that breech delivery was associated with a greater prevalence of intracranial hemorrhage and brachial plexus injury. 19n a study of 36 241 deliveries from the United States, occiput posterior position was associated with an 80% greater risk of skull fracture and Erb's paralysis compared with occiput anterior, and the risk increased twofold when an instrument was used. 18Others have shown that a prolonged second stage increases the risk of severe injury, 11 while women who bypass labor completely are less likely to have neonatal injuries. 15These factors should be considered in guidelines to prevent severe birth injury.
Birth weight may be an additional factor to consider.High birth weight is thought to increase the risk of severe injury, 20 but the extent of the association between forceps or vacuum and risk of injury in macrosomic infants has never been quantified.The association between high birth weight and severe birth injury is merely speculative at this point.In our analyses, operative vaginal delivery was a risk factor for severe injury regardless of birth weight.The associations were in fact weaker for macrosomic infants than infants with normal birth weight.Thus, the current belief that instrumentation is a greater risk for injury in macrosomic infants may be incorrect.Risks may in fact be greater for smaller infants.It may be that fetuses with higher birth weight are more robust and better able to tolerate instrumentation.Cesarean delivery after unsuccessful instrumentation may therefore be indicated when the fetus has normal or low weight.
2][23][24] Preterm neonates are particularly at risk of intraventricular hemorrhage, 21 with 15% of preterm newborns with very low birth weight having more severe grades of hemorrhage. 21A meta-analysis of 123 studies from the United States found that placental abruption was associated with up to 10 times the risk of cerebral palsy. 22ecent studies suggest that preeclampsia may increase the risk of intracranial hemorrhage. 23,24The extent to which these factors should be considered in guidelines to prevent severe injury will need to be examined in future research.
This study paints a portrait of the impact of successful and unsuccessful operative vaginal attempts on the risk of severe neonatal birth injury compared with spontaneous vaginal delivery.We assessed a range of injuries in a large cohort of neonates.Limitations are nevertheless present.We used administrative data that may be prone to misclassification due to coding errors that potentially attenuated the associations.We had no information on ethnicity, fetal presentation and position, or duration of labor, and cannot rule out the possibility of residual confounding.As this study was not a randomized trial, causal interpretations should be made with caution.We could not determine the underlying indication for operative vaginal or cesarean delivery, and did not examine hypoxic-ischemic injuries.The data reflect a population covered under publicly funded healthcare.Findings may not be generalizable to populations in different settings.
In this study of 1.1 million births in Canada, operative vaginal delivery, particularly combined use of forceps and vacuum, was strongly associated with risk of severe birth injury compared with spontaneous delivery.Unsuccessful operative vaginal attempts followed by cesarean delivery were also associated with severe birth injury, though to a lesser extent.The findings suggest that the impact of operative vaginal instrumentation on risk of severe birth injury may have been underestimated in previous studies and that cesarean delivery may be preferable when either forceps or vacuum fails.Current guidelines for prevention of severe neonatal birth injury may benefit from revision.

F I G U R E 1
Distribution of mode of delivery by type of severe birth injury.
Maternal and infant characteristics by mode of delivery.
Association of mode of delivery with risk of neonatal birth injury.
Association of operative vaginal delivery with type of severe birth injury.
Association of cesarean delivery with type of severe birth injury.
T A B L E 4 a Association for cesarean relative to spontaneous vaginal delivery, adjusted for maternal age, parity, infant sex, preterm birth, multiple birth, socioeconomic disadvantage, place of residence, and time period.Data are for births between 2009 and 2019.