Objective
Fetal size impacts on perinatal outcomes. We queried whether the fetal head, as the fetal part interfacing with the birth canal, might impact on obstetric outcomes more than birthweight (BW). We examined associations between neonatal head circumference (HC) and delivery mode and risk of perinatal complications as compared to high BW.
Study Design
This was an electronic medical records–based study of term singleton births (37-42 weeks’ gestation) from January 2010 through December 2012 (N = 24,780, 6343 primiparae). We assessed risks of unplanned cesarean or instrumental delivery and maternal and fetal complications in cases with HC or BW ≥95th centile (large HC, high BW) vs those with parameters <95th centile (normal). Newborns were stratified into 4 subgroups: normal HC/normal BW (reference, n = 22,548, primiparae 5862); normal HC/high BW (n = 817, P = 213); large HC/normal BW (n = 878, P = 265); and large HC/high BW (n = 537, P = 103). Multinomial multivariable regression provided adjusted odds ratio (aOR) while controlling for potential confounders.
Results
Infants with HC ≥95th centile (n = 1415) were delivered vaginally in 62% of cases, unplanned cesarean delivery 16%, and instrumental delivery 11.2%; 78.4% of infants with HC <95th centile were delivered vaginally, 7.8% unplanned cesarean, and 6.7% instrumental delivery. Odds ratio (OR) for unplanned cesarean was 2.58 (95% confidence interval [CI], 2.22–3.01) and for instrumental delivery OR was 2.13 (95% CI, 1.78–2.54). In contrast, in those with BW ≥95th centile (n = 1354) 80.3% delivered vaginally, 10.2% by unplanned cesarean (OR, 1.2; 95% CI, 1.01–1.44), and 3.4% instrumental delivery (OR, 0.46; 95% CI, 0.34–0.62) compared to infants with BW <95th centile: spontaneous vaginal delivery, 77.3%, unplanned cesarean 8.2%, instrumental 7.1%. Multinomial regression with normal HC/normal BW as reference group showed large HC/normal BW infants were more likely to be delivered by unplanned cesarean (aOR, 3.08; 95% CI, 2.52–3.75) and instrumental delivery (aOR, 3.03; 95% CI, 2.46–3.75). Associations were strengthened in primiparae. Normal HC/high BW was not associated with unplanned cesarean (aOR, 1.18; 95% CI, 0.91–1.54), while large HC/high BW was (aOR, 1.93; 95% CI, 1.47–2.52). Analysis of unplanned cesarean indications showed large HC infants had more failure to progress (27.7% vs 14.1%, P < .001), while smaller HC infants had more fetal distress (23.4% vs 16.9%, P < .05).
Conclusion
A large HC is more strongly associated with unplanned cesarean and instrumental delivery than high BW. Prospective studies are needed to test fetal HC as a predictive parameter for prelabor counseling of women with “big babies.”
Classic obstetrics describes the 3 P’s: passageway, passenger, and power of the uterus, and the dangers inherent in a mismatch. The fetal head represents the point of interface between “passenger” and “passageway.”
The literature of anthropology and human evolution addresses the issue of the obstetric dilemma of bipedalism and encephalization: ie, the tension at the point of meeting of the human fetal head and the bipedal female pelvis, but sparse attention has been paid in the obstetric literature to the impact of head circumference (HC) on obstetric outcomes. The literature on problems of delivery in “big babies” focuses primarily on fetal size in terms of weight. We queried how fetal head dimensions might impact the passage of the fetus through the birth canal and affect obstetric outcomes, primarily delivery mode. In our earlier work we showed that delivery of a fetus with large HC significantly increased the risk of intrapartum levator ani trauma.
Big babies and their mothers are at increased risk of prolonged second stage of labor, obstructed labor, unplanned cesarean delivery (UCD) or instrumental delivery, shoulder dystocia, brachial or facial nerve injuries, trauma to the clavicle and humerus, birth asphyxia, maternal postpartum hemorrhage, and trauma to the birth canal, bladder, or pelvic floor complex, with their attendant potential long-term effects. These clinical observations led to attempts to estimate fetal weight, clinically and in recent decades sonographically, to inform management decisions to reduce risks of fetal and maternal complications in labor and delivery of macrosomic fetuses. Professional guidelines devote considerable attention to big babies and the optimal delivery approach for them.
Definitions of macrosomia vary, and include term fetuses with estimated fetal weight (EFW) ≥4.5 kg (American Congress of Obstetricians and Gynecologists), or 4 kg or 5 kg, or based on the 97th centile of fetal weight for gestational age. However, overall, prenatal EFW is insufficiently sensitive in the prediction of macrosomia.
We posed the question whether the dimensions of the fetal head might be more important than fetal weight in determining the likelihood of cesarean or instrumental delivery. As a first step in answering this question, we designed an electronic medical records (EMR)-based study to test the notion whether HC impacts delivery mode more than baby weight. Here we aimed to determine the relationship between neonatal HC and mode of delivery and frequency of complications in labor, and to compare this with the relationship between birthweight (BW) and delivery mode and labor complications.
Materials and Methods
This was a cross-sectional study based on EMR performed in the Hadassah-Hebrew University Medical Centers Jerusalem, comprising 2 campuses of a tertiary care center. Our institutional ethical review board reviewed and approved the study (#0085-13-HMO; Feb. 14, 2013). All singleton term deliveries (37 +0 -42 +0 ) occurring from January 2010 through December 2012 were eligible for inclusion. Multiple births were excluded. Data were collected in blinded fashion by research staff who were not involved in any stage of perinatal care; midwives and obstetricians caring for laboring women and nursery physicians and nurses caring for their infants were unaware of the study. Data were extracted for maternal demographic and obstetric parameters including type of delivery and length of second stage of labor (defined as prolonged if >3 hours for primipara with epidural anesthesia, 2 hours without; 2 hours for multipara with epidural anesthesia, 1 hour without), maternal hemorrhage (defined as >500 mL or requiring transfusion of packed cells), and gestational diabetes mellitus, and neonatal parameters including BW, neonatal HC, infant sex, meconial fluid, umbilical cord (arterial) pH, 5-minute Apgar score, and neonatal intensive care unit admission. Infants with BW or HC ≤5th centile (n = 1780) were excluded from further analysis; our aim was to examine HC and BW to improve prelabor counseling and inform management decisions in big babies, and the subgroup at the opposite end of the spectrum includes a high proportion of pregnancies with complications such as preeclampsia, fetal growth restriction, and fetal anomalies, which elevate the risk for cesarean delivery and are managed under well-established protocols ( Figure 1 ).
BW was recorded in the delivery room; HC was measured in the newborn nursery 6-18 hours after delivery, with a flexible metal tape measure, passed around the baby’s head above the eyebrows anteriorly and at the posterior protuberance of the occipital bone posteriorly.
Outcomes were analyzed for primiparae and for the whole cohort. Infant sex was analyzed as a possible modifier. Statistical analyses were performed with SPSS 20 for Windows (IBM, Armonk, NY) and Excel (Microsoft Seattle, WA). Dichotomous variables were analyzed with the χ 2 test to determine proportions of each outcome differing from expected, and correlation between the variables. Fisher exact test was applied as appropriate in analyzing cells with small numbers of cases. Odds ratios (ORs) for 4 delivery modes (spontaneous vaginal, elective cesarean, instrumental [vacuum or forceps assisted], and unplanned cesarean) were determined for neonates with HC ≥95th centile and BW ≥95th centile, as compared to the expected risk in the whole cohort. Risks of the outcome parameters UCD or instrumental delivery were calculated for 7 percentile subgroups (5th, 10th, 25th, 50th, 75th, 90th, 95th) of HC and BW ( Figure 2 ).
Multinomial multivariable regression was used to obtain adjusted OR (aOR) of the mode of delivery for HC or BW ≥95th centile, with vaginal delivery as the reference group, while controlling for maternal age, primipara status, gestational age at delivery, and infant sex, as well as examining interactions between these variables. ORs and aORs are reported with 95% confidence intervals (CI).
An interaction term between HC and BW was found significant ( P < .001). However, not all neonates with large HC had high BW, and vice versa. To assess and compare the impact of HC and BW in neonates of differing proportions, a multinomial multivariable regression model for modes of delivery was analyzed according to 4 strata of HC and BW, where normal was defined as (>5th centile and <95th centile): normal HC/normal BW; large HC/normal BW; normal HC/high BW; and large HC/high BW.
To assess the feasibility of applying prenatal HC evaluation to prelabor counseling, fetal HC and EFW were compared to neonatal HC and BW to determine the degree of correlation between them. Ultrasound data were drawn from a subset of our study population undergoing ultrasound examination in our ultrasound units, including fetal biometry, within 3 days of delivery. Postnatal measurements were performed in identical fashion for all infants, and nursery caregivers were unaware of the prenatal ultrasound results. Prenatal sonographic HC and EFW were performed according to International Society of Ultrasound in Obstetrics and Gynecology guidelines. Sonographic EFW was derived from standard fetal biometry (abdominal circumference, femur length, biparietal diameter, and HC) with the Hadlock formula.
Bland-Altman plots and interclass correlation coefficients were applied to compare prenatal HC and EFW with postnatal measurements.
Results
A total of 28,168 term singleton deliveries occurred at our hospitals during the study period. Electronic data for HC and/or BW were missing in a small number of files: 1588 (5.6%) had no HC, 20 (0.07%) no BW, and 8 had neither parameter recorded. These cases were excluded from further analysis. The remaining cohort of 26,560 was analyzed for percentiles cutoffs; infants found to have HC and/or BW ≤5th centile were excluded from further analysis. A total of 24,780 singleton births therefore were included; 6343 were first deliveries ( Figure 1 ).
Background parameters of the study cohort and the 95th centile BW or HC subgroups are shown in Table 1 . One quarter of the patients in the cohort were primiparae (25.6%), while 9.7% were grand-multiparae. Rates of gestational diabetes mellitus did not differ significantly among the groups. The rates and ORs of delivery mode outcomes (spontaneous vaginal delivery being the reference group) in large HC and high BW babies, among the whole cohort and primiparae, are shown in Table 2 .
| Demographic | Cohort a | ≥95th centile HC | ≥95th centile BW | P value b |
|---|---|---|---|---|
| n = 24,780 | n = 1415 | n = 1354 | ||
| Maternal age, mean (± SD) | 30.14 (5.56) | 30.93 (5.56) | 31.02 (5.36) | < .001 |
| Parity, n (%) | ||||
| Primipara | 6343 (25.6) | 368 (26) | 216 (16) | < .001 |
| Multipara | 16,037 (64.7) | 857 (60.6) | 893 (66) | |
| Grandmultipara | 2400 (9.7) | 190 (13.4) | 245 (18.1) | |
| GA at delivery, median (range) | 39 (37–42) | 40 (37–42) | 40 (37–42) | < .001 |
| Infant sex, n (% female) | 11,811 (47.7) | 315 (22.3) | 448 (33.1) | < .001 |
| Current smoker, n (% yes) | 851 (3.5) | 23 (1.7) | 25 (1.9) | < .001 |
| GDM, n (%) | 824 (3.4) | 54 (3.8) | 53 (3.9) | NS |
| HC, mean (range) | 34.46 (32.3–41) | 36.74 (36.2–41) | 35.96 (33–40) | < .001 |
| BW, mean (range) | 3351.1 (2600–5250) | 3899.8 (2664–5250) | 4214.42 (4018–5250) | < .001 |
a Excluding infants with HC and/or BW measurements ≤5th centile
b P values were calculated for χ 2 test for dichotomous variables for subgroup with HC ≥95th centile or BW ≥95th centile vs the cases with HC or BW measuring >5th and <95th centiles.
| Variable | % of all deliveries | Head circumference | Birthweight | ||||
|---|---|---|---|---|---|---|---|
| Rate of delivery mode in ≥95th centile, % | Rate of delivery mode in <95th centile, % | OR (95% CI) (≥95th vs <95th centile) | Rate of delivery mode in ≥95th centile, % | Rate of delivery mode in <95th centile, % | OR (95% CI) (≥95th vs <95th centile) | ||
| Spontaneous vaginal delivery a | |||||||
| Cohort | 77.4 | 62 | 78.4 | Reference category | 80.3 | 77.3 | Reference category |
| Primiparae | 63.4 | 38 | 63.4 | Reference category | 56.5 | 62.2 | Reference category |
| Instrumental | |||||||
| Cohort | 6.9 | 11.2 | 6.7 | 2.13 (1.78–2.54) | 3.4 | 7.1 | 0.46 (0.34–0.62) |
| Primiparae | 17.1 | 23.1 | 16.8 | 2.29 (1.74–3.03) | 10.6 | 17.4 | 0.68 (0.43–1.06) |
| Elective cesarean | |||||||
| Cohort | 7.4 | 10.7 | 7.2 | 1.9 (1.59–2.27) | 6.1 | 7.4 | 0.79 (0.63–0.99) |
| Primiparae | 4.6 | 6.8 | 4.5 | 2.53 (1.62–3.94) | 6.9 | 4.5 | 1.69 (0.97–2.92) |
| Unplanned cesarean | |||||||
| Cohort | 8.3 | 16 | 7.8 | 2.58 (2.22–3.01) | 10.2 | 8.2 | 1.2 (1.01–1.44) |
| Primiparae | 16.3 | 32.1 | 15.3 | 3.5 (2.7–4.51) | 25.9 | 15.9 | 1.79 (1.3–2.48) |
Rate of vaginal delivery was 78.4% when HC <95th centile vs 62% in HC ≥95th centile. Risk of UCD was 7.8% when HC <95th centile and increased to 16% when HC ≥95th centile. Risk of instrumental delivery was 6.7% and 11.2% in babies with HC below and above the 95th centile, respectively. (Results for infants ≥97.5 centile are shown in Supplemental Table 1 ; Appendix .) The OR for UCD was 2.58 (95% CI, 2.22–3.01) and for instrumental delivery OR 2.13 (95% CI, 1.78–2.54) for babies with HC ≥95th centile as compared to those with HC in normal range.
Babies with BW ≥95th centile delivered vaginally in 80.3% of cases, while 10.2% were delivered by UCD and 3.4% by instrumental delivery. BW ≥95th centile showed weaker association with UCD (OR, 1.2; 95% CI, 1.01–1.44) than HC, and was associated with significantly lower odds of instrumental delivery (OR, 0.46; 95% CI, 0.34–0.62). ( Supplemental Table 2 compares ORs in infants at 95 and 97.5 centile subgroups.)
Multinomial regression modeling controlling for maternal age, gestational age at delivery, sex of the fetus, and primipara status showed that HC ≥95th centile more than doubled the risk of UCD (aOR, 2.65; 95% CI, 2.22–3.17) while BW ≥95th centile had no significant effect (OR, 0.91; 95% CI, 0.73–1.12) ( Table 3 ). Interaction terms between infant sex and HC and BW were nonsignificant (data not shown).
| Variable | Vaginal delivery (reference group) | Instrumental delivery aOR (95% CI) | Elective cesarean aOR (95% CI) | Unplanned cesarean aOR (95% CI) |
|---|---|---|---|---|
| BW and HC stratum | ||||
| HC ≥95th centile | 1 | 2.67 (2.18–3.27) | 3.23 (2.57–4.06) | 2.65 (2.22–3.17) |
| BW ≥95th centile | 1 | 0.34 (0.24–0.47) | 1.14 (0.86–1.51) | 0.9 (0.73–1.12) |
| Maternal age (continuous, years) | 1 | 1.03 (1.02–1.04) | 1.12 (1.1–1.13) | 1.08 (1.07–1.09) |
| GA (continuous, 37-42 wks) | 1 | 1.12 (1.07–1.17) | 0.32 (0.3–0.34) | 0.95 (0.92–0.99) |
| Infant sex | ||||
| Female | 1 | – | – | – |
| Male | 1 | 1.3 (1.17–1.44) | 0.85 (0.76–0.94) | 1.33 (1.2–1.46) |
| Parity | ||||
| Parity>1 | 1 | – | – | – |
| Primipara status | ||||
| Primipara | 1 | 7.49 (6.67–8.41) | 1.38 (1.19–1.6) | 5.93 (5.34–6.6) |
Large HC does not always accompany high BW ( Figure 3 ): only in 24% of big fetuses, ie, those with HC or BW ≥95th centile, did both parameters occur together. Some 2.2% of the whole study cohort had both HC and BW ≥95th centile. An interaction term between HC and BW was found significant ( P < .001) in multinomial multivariable regression analysis for mode of delivery adjusted to the above parameters (model not shown). To investigate this further we analyzed modes of delivery according to 4 strata, where normal was defined as (>5th centile and <95th centile): normal HC/normal BW; large HC/normal BW; normal HC/high BW; and large HC/high BW ( Table 4 ). Rates of spontaneous vaginal delivery in the subgroups were highest in the normal HC/normal BW and normal HC/high BW subgroups: 78.1% and 85.1%, respectively. Large HC/normal BW babies were delivered vaginally in only 55.4% of cases, and large HC/high BW babies, 73%.
| Variable | Vaginal delivery rate, % (reference group) | Instrumental delivery | Elective cesarean | Unplanned cesarean | |||
|---|---|---|---|---|---|---|---|
| Rate, % | aOR (95% CI) | Rate, % | aOR (95% CI) | Rate, % | aOR (95% CI) | ||
| BW and HC stratum (cohort) a | |||||||
| Normal HC/Normal BW (n = 22,548) | 78.1 | 6.8 | – | 7.3 | – | 7.8 | – |
| Large HC/Normal BW (n = 878) | 55.4 | 15.8 | 3.03 (2.46–3.75) | 11.3 | 3.26 (2.52–4.21) | 17.5 | 3.08 (2.52–3.75) |
| Normal HC/High BW (n = 817) | 85.1 | 3.2 | 0.52 (0.35–0.78) | 3.7 | 1.08 (0.73–1.6) | 8.1 | 1.18 (0.91–1.54) |
| Large HC/High BW (n = 537) | 73 | 3.7 | 0.59 (0.37–0.94) | 9.9 | 3.71 (2.65–5.2) | 13.4 | 1.93 (1.47–2.52) |
| BW and HC stratum (primiparae) b | |||||||
| Normal HC/Normal BW (n = 5862) | 63.4 | 16.9 | – | 4.5 | – | 15.2 | – |
| Large HC/Normal BW (n = 265) | 34.4 | 27.5 | 2.67 (1.94–3.68) | 6 | 4.32 (2.37–7.86) | 32.1 | 3.3 (2.41–4.53) |
| Normal HC/High BW (n = 213) | 64.6 | 9.7 | 0.52 (0.28–0.99) | 5.3 | 3.69 (1.51–8.99) | 20.4 | 1.28 (0.79–2.09) |
| Large HC/High BW (n = 103) | 47.6 | 11.7 | 0.81 (0.43–1.53) | 8.7 | 6.36 (2.89–13.99) | 32 | 2.36 (1.48–3.75) |
a Adjusted for maternal age, GA at delivery, infant sex, and parity
Normal HC/normal BW babies underwent instrumental delivery in 6.8% of cases, as compared to 15.8% of large HC/normal BW cases. This is in contrast to a low rate in normal HC/high BW and large HC/high BW infants, who were delivered by vacuum extraction 3.2 and 3.7% of the time, respectively.
Normal HC/normal BW babies were delivered by UCD in 7.8% of cases, and normal HC/high BW cases, 8.1%. In contrast, 17.5% of large HC/normal BW cases were delivered by UCD, and 13.4% of large HC/high BW infants. Risk of interventional delivery was higher in all subgroups among primiparae.
Multinomial multivariable regression applying these HC/BW strata to mode of delivery and adjusted for the parameters above showed that the risk of UCD for the large HC/normal BW group was 3 times (aOR, 3.08; 95% CI, 2.52–3.75) that of the normal HC/normal BW group. The risk of instrumental delivery was aOR 3.03 (95% CI, 2.46–3.75). Among primiparae, risk of UCD rose to 3.3 (95% CI, 2.41–4.53) while risk of instrumental delivery decreased slightly to 2.67 (95% CI, 1.94–3.68). In the 2 high BW subgroups (with or without large HC) the risk of UCD was nearly doubled among the large HC/high BW infants (aOR, 1.93; 95% CI, 1.47–2.52), but not significantly increased in normal HC/high BW cases (aOR, 1.18; 95% CI, 0.91–1.54). Notably, high BW cases with large or normal HC were significantly less likely to undergo instrumental delivery (aOR, 0.59; 95% CI, 0.37–0.94 and aOR, 0.52; 95% CI, 0.35–0.78, respectively). Table 4 summarizes rates and aORs of delivery modes in the 4 strata. Among primiparae, risk of UCD among high BW infants was 2.36 (95% CI, 1.48–3.75) when HC ≥95th centile, but was not elevated when HC was normal (aOR, 1.28; 95% CI 0.79–2.09); large HC/normal BW was associated with increased risk of instrumental delivery (aOR, 2.67; 95% CI, 1.94–3.68). The normal HC/high BW stratum was at reduced risk for instrumental delivery (aOR, 0.52; 95% CI, 0.28–0.99) while large HC/high BW stratum showed a nonsignificant reduction in risk of instrumental delivery (aOR, 0.81; 95% CI, 0.43–1.53).
In analysis of the main indications for cesarean delivery (other than elective indications) we found that infants with large HC were more likely to undergo cesarean delivery for failure to progress (27.7% vs 14.1%, P < .001), while smaller infants were more likely to be delivered by cesarean for fetal distress (23.4% vs 16.9%, P < .05) (data not shown).
Large HC impacted other outcome parameters. HC ≥95th centile was significantly associated with prolonged second stage of labor and maternal hemorrhage. Infants with HC ≥95th centile were more likely to be hospitalized in the special care nursery or neonatal intensive care unit, to have Apgar score ≤7, and to have umbilical cord pH ≤7.1 ( Table 5 , Supplemental Tables 1 and 2 ).
| Variable | Total | Head circumference | Birthweight | ||
|---|---|---|---|---|---|
| <95th centile | ≥95th centile | <95th centile | ≥95th centile | ||
| Prolonged second stage | |||||
| Cohort | 1325 (5.3%) | 1062 (5.3%) | 131 (12.2%) | 1128 (5.8%) | 65 (5.8%) |
| Primiparae | 755 (14.3%) | 657 (13.5%) | 80 (29.3%) | 722 (14.1%) | 33 (19.6%) |
| Maternal postpartum hemorrhage | |||||
| Cohort | 2695 (10.9%) | 2460 (10.5%) | 235 (16.6%) | 2481 (10.6%) | 214 (15.8%) |
| Primiparae | 962 (15.2%) | 871 (14.6%) | 91 (24.7%) | 897 (14.6%) | 65 (30.1%) |
| Meconial fluid | |||||
| Cohort | 1763 (7.1%) | 1641 (7%) | 122 (8.6%) | 1652 (7.1%) | 111 (8.2%) |
| Primiparae | 574 (9%) | 534 (8.9%) | 40 (10.9%) | 550 (9%) | 24 (11.1%) |
| Apgar <7 | |||||
| Cohort | 29 (0.1%) | 25 (0.1%) | 4 (0.3%) | 27 (0.1%) | 2 (0.1%) |
| Primiparae | 17 (0.3%) | 13 (0.2%) | 4 (1.1%) | 16 (0.3%) | 1 (0.5%) |
| pH ≤7.1 | |||||
| Cohort | 182 (0.9%) | 164 (0.9%) | 18 (1.5%) | 175 (1%) | 7 (0.7%) |
| Primiparae | 86 (1.7%) | 76 (1.6%) | 10 (3.2%) | 83 (1.7%) | 3 (1.7%) |
| NICU a | |||||
| Cohort | 57 (0.2%) | 48 (0.2%) | 9 (0.6%) | 53 (0.2%) | 4 (0.3%) |
| Primiparae | 25 (0.4%) | 20 (0.3%) | 5 (1.4%) | 25 (0.4%) | 0 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
