Schlüsselwörter
vakuumassistierte Geburt - Vakuumextraktion - subgaleales Hämatom - Kephalhämatom - Kopfverletzungen - niedriges Geburtsgewicht - zunehmendes fetales Gewicht - Komplikationen der Vakuumextraktion
Keywords
vacuum-assisted delivery - vacuum extraction - subgaleal hematoma - cephalohematoma - head injuries - low birthweight - increasing fetal weight - vacuum complications
Abbreviations
BW:
birthweight
CD:
cesarean delivery
GA:
gestational age
LBW:
low birthweight
NICU:
neonatal intensive care unit
NRFHR:
non-reassuring fetal heart rate
OA:
occiput anterior
SDH:
subdural hematoma
SGH:
subgaleal hematoma
VE:
vacuum extraction
Introduction
Vacuum extractions (VE) are used in 10–13% of deliveries in the UK [1]. In the USA, rates are decreasing [2]. Despite different attitudes, there is a worldwide consensus that early preterm delivery (gestational age [GA] < 34 w) is a relative
contraindication for VE, due to an increased rate of intraventricular hemorrhage [2]
[3]. However, this consensus is based mainly on empirical recommendations, while data regarding this GA and a safe lower age limit have yet to be established [4]
[5]. Moreover, existing data do not differentiate clearly between preterm delivery and
low birthweight (LBW). Therefore, it is uncertain whether the increased neonatal risks of VE are related to the preterm delivery or to the accompanying LBW.
VE is a risk factor for several neonatal injuries related to birth trauma [6]
[7].
It has also been specifically linked to neonatal cephalohematoma, subgaleal hematoma (SGH), intracranial hemorrhage and skull fractures [4]
[8]
[9].
Currently, data linking a higher rate of birth trauma to decreasing GA or neonatal birthweight (BW) are controversial. Two small studies evaluated outcomes in preterms with LBW, comparing VE
with unassisted vaginal delivery. Both demonstrated VE to be relatively safe for small neonates, finding no increase in neonatal morbidity or intraventricular hemorrhage [10]
[11].
A larger study evaluated the risks of VE compared to other modes of delivery in preterms during labor and found higher odds ratios for intra- and extracranial hemorrhages with VE [4].
All the above studies compared VE with other modes of delivery at a certain weight threshold. This comparison is limited because VE is always associated with higher rates of neonatal
complications. Therefore, to evaluate the true effect of LBW, a better comparison would be between VE performed at different neonatal weights. This comparison was conducted by Aviram et al.,
who compared neonatal injuries during VE using a metal vacuum cup. Birth trauma was similar between neonates with birthweights below and above 2500 grams [5].
Since metal cups are associated with more neonatal birth trauma, as compared to non-metal cups, many physicians favor the latter [12]
[13]
[14]
[15]. Common non-metal cups include the Kiwi Omnicup (Clinical Innovations, Murray, Utah, USA) and the mushroom-shaped Ventouse-Mityvac vacuum-assisted delivery system
(CooperSurgical, Inc., Trumbull, CT, USA). Both systems consist of rigid, mushroom-shaped disposable cups. The main differences between them are the vacuum mechanism (handheld pump vs.
conventional vacuum) and the traction pole (flexible in the Kiwi and rigid in the Mityvac). Previous studies comparing these two cups found no differences in adverse neonatal outcomes [16].
We aimed to evaluate the impact of VE on neonatal head injuries related to birth trauma in neonates with LBW (< 2500 g), using a non-metal vacuum cup.
Materials and Methods
This is a retrospective cohort study, including singleton pregnancies with no known genetic or structural anomalies which were delivered by VE using a non-metal cup, from January 2014 to
August 2019 in a single academic institution. All VE were performed at GA ≥ 34 weeks.
To ensure that the reported birth injuries were associated with VE only, we excluded a single case of failed vacuum extraction with a subsequent forceps delivery, and included cases of either
successful VE or failed VE followed by cesarean delivery.
Each VE was attended by a senior physician (a graduate of a 6-year OBGYN residency) who performed a full evaluation before the procedure and ensured that the conditions met the American
College of Obstetricians and Gynecologists (ACOG) guidelines [17]. The procedure itself was performed by an OBGYN resident supervised by
a senior physician, or by the senior physician himself when residents were absent or had difficulties during the procedure. This protocol was maintained irrespective of fetal estimated weight
or gestational age during delivery. A pediatrician was present at every VE. All VE were performed using a non-metal cup (Ventouse-Mityvac or Kiwi Omnicup), according to physician’s preference.
Cup size was identical for both cups used – 50 mm. Metal cups were not used in our institution. The procedure was aborted after 3 cup detachments or when procedure duration reached more than
30 minutes, according to the guidelines of the Israel Society of Obstetrics and Gynecology [18]. After delivery, the physician completed
a detailed report detailing the assessment of the labor pattern before and during the procedure.
We divided our cohort into 2 groups according to neonatal birthweights: neonates weighing < 2500 g or ≥ 2500 g. The second group was divided further into 3 subgroups: 2500–2999 g,
3000–3499 g and ≥ 3500 g. Primary and secondary outcome rates were also calculated for the proportion of newborns weighing ≥ 4000 g.
First, we compared the < 2500 g neonates with the ≥ 2500 g neonates. Subsequently, we compared the < 2500 g neonates with each of the 3 subgroups, individually. Primary and secondary
outcome rates were also calculated for the proportion of newborns weighing ≥ 4000 g in group 4.
Since neonatal BW should be correlated to GA, we also performed a comparison between neonates who were small, average, and large for gestational age (SGA, AGA, LGA respectively), in order to
evaluate the neonatal outcomes in each group. SGA, AGA and LGA were diagnosed according to local birthweight charts [19].
The primary composite neonatal outcome was defined as neonatal head injuries related to birth trauma, including one or more of the following: SGH, cephalohematomas, SDH and intracranial
hemorrhages. We also calculated the rates of SGH and cephalohematomas for each group.
Secondary outcomes were neonatal intensive care unit (NICU) hospitalization rates and a secondary composite neonatal outcome including one or more of the following neonatal outcomes not
related to birth trauma: Apgar scores, respiratory distress, need for mechanical ventilation, meconium aspiration, hypoxic ischemic encephalopathy, convulsions, and neonatal sepsis.
Data were retrieved using documentation in the delivery room and in the neonatal unit. All medical records were reviewed to complete missing data.
Data included maternal demographics (age, gravidity, parity, body mass index [BMI], GA at delivery, pregestational or gestational diabetes, and a history of cesarean delivery [CD]). Labor and
delivery characteristics collected were use of epidural analgesia; intrapartum fever; duration of first, second and third stages of labor; indication for VE; fetal head position and station
before VE; cup type; vacuum duration and cup detachments.
Data collected regarding neonatal outcomes included fetal weight, rates of SGA, AGA and LGA neonates, NICU hospitalization, and head injuries related to birth trauma (SGH, cephalohematoma,
subdural hematoma (SDH) and intracranial hemorrhage). Data regarding neonatal outcomes not related to birth trauma included Apgar scores, respiratory distress and need for mechanical
ventilation, meconium aspiration, hypoxic-ischemic encephalopathy, convulsions, and neonatal sepsis.
Gestational age at delivery was confirmed by first trimester crown rump length (CRL). Gestational diabetes was defined according to ACOG guidelines [20], and pregestational diabetes was defined when mentioned in the patient’s medical records. Intrapartum fever was defined as two or more measurements of a temperature
≥ 38.0 degree Celsius during labor and up to 24 hours post partum. Fetal head position before VE was determined by both a vaginal examination and intrapartum sonography, and fetal head station
was determined by a vaginal examination.
Indications for VE were divided into:
-
Non-reassuring fetal heart rate (NRFHR) defined according to the NICHD three-tier system [21],
-
prolonged second stage, defined according to ACOG guidelines [22], and
-
maternal indications, including exhaustion or medical background requiring shortening of second stage.
For historical reasons, fetal head station was defined by thirds from ischial spines −3 to +3, and was divided into 3 groups – mid-pelvis: S+1, low: S+2, and outlet: S+3 and below. VE was
only performed when the fetal head was below the ischial spines according to ACOG guidelines.
Neonatal diagnoses were determined by the senior pediatrician present at VE and during neonatal hospitalization, according to international standards and relevant imaging (ultrasound and
magnetic resonance imaging).
Data analysis
Distribution of continuous variables was evaluated using the Shapiro-Wilk test. Variables with a normal distribution were presented as mean + SD, and were compared using t-test. Categorical
data were compared using chi-square or Fisher’s exact test (as appropriate). Multivariable logistic regression was also performed using variables that were found to be significantly
different between the groups in the chi-square analysis (vacuum cup type, procedure duration, intrapartum fever, head station at vacuum extraction, and duration of second stage of labor).
Adjusted odds ratios and their 95% confidence intervals were calculated. A probability value of p < 0.05 was considered significant. Analyses were performed using SPSS 26 software (IBM,
Armonk, NY, USA).
Results
During the study period, 41052 women were delivered in our institution, of whom 3377 (8.2%) had a VE and met the inclusion criteria ([Fig. 1]).
Fig. 1
Flowchart describing the study population.
Overall, 206 (6.1%) neonates were included in the LBW group (< 2500 g, group 1), 884 (26.2%) in the 2500–2999 g group (group 2), 1451 (43.0%) in the 3000–3499 g group (group 3) and 836
(24.8%) in the ≥ 3500 g group (group 4). Within group 4, 108 neonates (3.2% of the total cohort) were born with a BW ≥ 4000 g.
Birth trauma-related neonatal outcomes
The lowest rates of SGH occurred in neonates with LBW < 2500 g (0.5%) and the rate increased with every additional 500 g of neonatal weight (3.5%, 4.4% and 8.0% in the 2500–2999 g,
3000–3499 g, ≥ 3500 g groups respectively, and 16.7% for the percentage of neonates weighing ≥ 4000 g, p = 0.001, [Table 1]).
Table 1
Head injuries related to birth trauma stratified by neonatal birthweight.
|
Variable
|
< 2500 g
(n = 206)
|
≥ 2500 g
(n = 3171)
|
P value
|
2500–2999 g
(n = 884)
|
P value
|
3000–3499 g
(n = 1451)
|
P value
|
≥ 3500 g
(n = 836)
|
P value
|
≥ 4000 g
(n = 108)
|
P value
|
|
Head trauma (n, %)
|
Subgaleal hematoma
|
1
|
0.5%
|
162
|
5.1%
|
0.001
|
31
|
3.5%
|
0.019
|
64
|
4.4%
|
0.003
|
67
|
8.0%
|
0.012
|
18
|
16.7%
|
< 0.001
|
|
Cephalohematoma
|
1
|
0.5%
|
102
|
3.2%
|
0.020
|
23
|
2.6%
|
0.066
|
48
|
3.3%
|
0.025
|
31
|
3.7%
|
0.012
|
3
|
2.8%
|
0.119
|
|
Subdural hematoma
|
0
|
0.0%
|
2
|
0.1%
|
1.000
|
1
|
0.1%
|
1.000
|
1
|
0.1%
|
1.000
|
0
|
0.0%
|
-
|
0
|
0.0%
|
-
|
|
Intracranial hemorrhage
|
0
|
0.0%
|
2
|
0.1%
|
1.000
|
1
|
0.1%
|
1.000
|
1
|
0.1%
|
1.000
|
0
|
0.0%
|
-
|
0
|
0.0%
|
-
|
|
Total head trauma
|
2
|
1.0%
|
266
|
8.4%
|
< 0.001
|
55
|
6.2%
|
0.001
|
113
|
7.8%
|
< 0.001
|
98
|
11.7%
|
< 0.001
|
21
|
19.4%
|
< 0.001
|
Cephalohematomas occurred significantly less often among LBW neonates (0.5% in < 2500 g BW) and increased with every additional 500 g of BW (2.6%, 3.3% and 3.7% in the 2500–2999 g,
3000–3499 g, and ≥ 3500 g groups, respectively, p = 0.020). The proportion of neonates weighing ≥ 4000 g had relatively more cephalohematomas compared to LBW neonates, but this difference
did not cross the level of significance (2.8% vs 0.5%, p = 0.119).
SDH and intracranial hemorrhage were rare, and occurred only in two neonates weighing 2500–3499 g.
The primary outcome of head injuries related to birth trauma, including one or more of the injuries mentioned above, increased significantly with increasing neonatal BW, from 1% in neonates
weighing < 2500 g, vs. 6.2%, 7.8% and 11.7% in the 2500–2999 g, 3000–3499 g, ≥ 3500 g groups, respectively, and 19.4% in the group of neonates weighing ≥ 4000 g, p < 0.001 ([Table 1]).
Logistic regression found three independent, significant risk factors for neonatal head injuries: higher neonatal weight (≥ 2500 g compared to < 2500 g) – with an adjusted odds ratio
(OR) of 11.30 (95% confidence interval = 1.57, 81.50, p = 0.016), longer procedure duration (OR 1.11 for every added minute, CI = 1.07, 1.15, p < 0.001) and use of the Ventouse-Mityvac
cup (OR 1.79, CI = 1.33, 2.41, p < 0.001). Duration of second stage labor, head station at the time of VE, and intrapartum fever were not significant in terms of neonatal head injuries
([Table 2]). The ORs for neonatal head injuries with increasing neonatal BW were 8.12, 10.88, and 13.52 for the 2500–2999 g, 3000–3499 g, and
≥ 3500 g groups, respectively, and 17.44 for BW ≥ 4000 g, (p = 0.008), ([Table 3]).
Table 2
Multivariable logistic regression analysis of factors associated with neonatal head injuries.
|
Variable
|
P value
|
Odds ratio
|
95% Confidence interval
|
|
Variables that differed significantly between BW groups in univariate analysis were included in the multivariable logistic regression analysis. These included: neonatal weight,
vacuum cup type, procedure duration, head station at VE, duration of second stage, intrapartum fever.
|
|
Neonatal weight (< 2500 g vs. ≥ 2500 g)
|
0.016
|
11.30
|
1.57, 81.50
|
|
Vacuum cup type (VM)
|
< 0.001
|
1.79
|
1.33, 2.41
|
|
Procedure duration (every added minute)
|
< 0.001
|
1.11
|
1.07, 1.15
|
|
Head station at vacuum extraction
|
0.343
|
2.00
|
0.48, 8.36
|
|
Duration of second stage
|
0.862
|
1.00
|
0.99, 1.01
|
|
Intrapartum fever
|
0.930
|
1.02
|
0.63, 1.67
|
Table 3
Adjusted odds ratio for neonatal head injuries by increasing BW.
|
Variable
|
P value
|
Odds ratio
|
95% Confidence interval
|
|
< 2500 g
|
–
|
1
|
–
|
|
2500–2999 g
|
0.040
|
8.12
|
1.10, 59.78
|
|
3000–3499 g
|
0.018
|
10.88
|
1.49, 79.28
|
|
≥ 3500 g
|
0.011
|
13.52
|
1.84, 99.38
|
|
≥ 4000 g
|
0.008
|
17.44
|
2.10, 144.49
|
General neonatal outcomes
NICU hospitalization rates were highest among neonates weighing < 2500 g (10.2%) compared to the other groups (3.1%, 1.7% and 3.3% in the 2500–2999 g, 3000–3499 g, and ≥ 3500 g group,
respectively, p < 0.001) and were comparable between neonates weighing < 2500 g and neonates weighing ≥ 4000 g (10.1% vs. 4.6% p = 0.130; [Table 4]).
Table 4
General neonatal outcomes stratified by neonatal birthweight.
|
Variable
(n, %)
|
< 2500 g
(n = 206)
|
≥ 2500 g
(n = 3171)
|
P value
|
2500–2999 g
(n = 884)
|
P value
|
3000–3499 g
(n = 1451)
|
P value
|
≥ 3500 g
(n = 836)
|
P value
|
≥ 4000 g
(n = 108)
|
P value
|
|
Apgar 5 min ≤ 7
|
3
|
1.5%
|
25
|
0.8%
|
0.242
|
4
|
0.5%
|
0.104
|
11
|
0.8%
|
0.403
|
10
|
1.2%
|
0.729
|
1
|
0.9%
|
1.000
|
|
Convulsions
|
0
|
0.0%
|
2
|
0.1%
|
1.000
|
0
|
0.0%
|
–
|
1
|
0.1%
|
1.000
|
1
|
0.1%
|
1.000
|
0
|
0.0%
|
–
|
|
Respiratory distress
|
2
|
1.0%
|
20
|
0.6%
|
0.392
|
4
|
0.5%
|
0.318
|
7
|
0.5%
|
0.310
|
9
|
1.1%
|
1.000
|
4
|
3.7%
|
0.187
|
|
Need for mechanical ventilation
|
2
|
1.0%
|
14
|
0.4%
|
0.255
|
3
|
0.3%
|
0.240
|
6
|
0.4%
|
0.261
|
5
|
0.6%
|
0.630
|
0
|
0.0%
|
–
|
|
Meconium aspiration
|
0
|
0.0%
|
5
|
0.2%
|
1.000
|
2
|
0.2%
|
1.000
|
2
|
0.1%
|
1.000
|
1
|
0.1%
|
1.000
|
0
|
0.0%
|
–
|
|
Hypoxic ischemic encephalopathy
|
0
|
0.0%
|
3
|
0.1%
|
1.000
|
0
|
0.0%
|
–
|
2
|
0.1%
|
1.000
|
1
|
0.1%
|
1.000
|
1
|
0.9%
|
0.344
|
|
Neonatal sepsis
|
0
|
0.0%
|
5
|
0.2%
|
1.000
|
1
|
0.1%
|
1.000
|
2
|
0.1%
|
1.000
|
2
|
0.2%
|
1.000
|
0
|
0.0%
|
–
|
|
Secondary composite outcome
|
7
|
3.4%
|
81
|
2.6%
|
0.461
|
17
|
1.9%
|
0.194
|
35
|
2.4%
|
0.400
|
29
|
3.5%
|
0.960
|
5
|
4.6%
|
0.555
|
|
NICU admission
|
21
|
10.2%
|
79
|
2.5%
|
< 0.001
|
27
|
3.1%
|
< 0.001
|
24
|
1.7%
|
< 0.001
|
28
|
3.3%
|
< 0.001
|
5
|
4.6%
|
0.130
|
No differences were noted between the groups regarding the Apgar scores ≤ 7 at 5 minutes, respiratory distress, need for mechanical ventilation, meconium aspiration, hypoxic-ischemic
encephalopathy and neonatal sepsis, or the total secondary composite neonatal outcome.
Demographic characteristics
No differences were found between the groups regarding the demographic characteristics, except for GA at delivery, which, as was to be expected based on the study design, increased with
increasing birthweight for each weight group. ([Table 5]).
Table 5
Baseline characteristics in relation to neonatal birthweight.
|
Variable
|
< 2500 g
(n = 206)
|
≥ 2500 g
(n = 3171)
|
P value
|
2500–2999 g
(n = 884)
|
P value
|
3000–3499 g
(n = 1451)
|
P value
|
≥ 3500 g
(n = 836)
|
P value
|
|
GDM = gestational diabetes mellitus
|
|
Maternal age (years), mean ± SD
|
30.2 ± 5.2
|
30.2 ± 5.4
|
0.937
|
30.4 ± 5.2
|
0.654
|
29.9 ± 5.1
|
0.491
|
30.6 ± 5.3
|
0.363
|
|
Parity, mean ± SD
|
0.4 ± 0.9
|
0.4 ± 0.8
|
0.584
|
0.3 ± 0.7
|
0.293
|
0.41 ± 0.8
|
0.899
|
0.6 ± 1.0
|
0.008
|
|
Nulliparity (n, %)
|
166
|
80.6%
|
2478
|
78.8%
|
0.411
|
736
|
83.3%
|
0.360
|
1153
|
79.5%
|
0.709
|
589
|
70.5%
|
0.004
|
|
Previous cesarean delivery (n, %)
|
14
|
6.8%
|
248
|
7.8%
|
0.593
|
64
|
7.2%
|
0.824
|
119
|
8.2%
|
0.487
|
65
|
7.8%
|
0.631
|
|
Gestational age [days, (weeks)], mean ± SD
|
262.1 (37.4) ± 11.1 d
|
277.3 (39.6) ± 9.3 d
|
< 0.001
|
274.2 (39.2) ± 9.4 d
|
< 0.001
|
278.7 (39.9) ± 7.4 d
|
< 0.001
|
282.0 (40.4) ± 6.5 d
|
< 0.001
|
|
GDM (n, %)
|
17
|
8.3%
|
228
|
7.2%
|
0.569
|
60
|
6.8%
|
0.460
|
109
|
7.5%
|
0.708
|
59
|
7.1%
|
0.555
|
|
Maternal BMI, mean ± SD
|
25.1 ± 5.2
|
23.2 ± 4.9
|
0.077
|
21.8 ± 4.7
|
0.005
|
23.4 ± 5.2
|
0.145
|
23.9 ± 4.6
|
0.259
|
|
Maternal obesity (BMI > 30) (n, %)
|
4
|
1.9%
|
37
|
1.2%
|
0.313
|
4
|
0.5%
|
0.046
|
22
|
1.5%
|
0.555
|
11
|
1.3%
|
0.513
|
Maternal BMI were generally similar between groups, except it was lower among mothers who delivered infants in the 2500–2999 g group (21.8 ± 4.7 vs. 25.1 ± 5.2, p = 0.005). Maternal obesity
rates (BMI > 30 kg/m2) were also lower in that group (0.5% vs. 1.9%, p = 0.046).
Nulliparity rates were similar between all groups, excluding women who delivered neonates ≥ 3500 g in whom nulliparity percentage was lower (70.5% vs. 80.6%, p = 0.004).
Labor and delivery characteristics
The rates of intrapartum fever were significantly lower in women delivering LBW neonates and increased with increasing neonatal BW (4.9% in < 2500 g group vs. 5.7%, 8.9% and 13.0% in the
2500–2999 g, 3000–3499 g and ≥ 3500 g groups respectively, p = 0.038; [Table 6]).
Table 6
Labor and delivery characteristics in relation to neonatal birthweight.
|
Variable
|
< 2500 g
(n = 206)
|
≥ 2500 g
(n = 3171)
|
P value
|
2500–2999 g
(n = 884)
|
P value
|
3000–3499 g
(n = 1451)
|
P value
|
≥ 3500 g
(n = 836)
|
P value
|
|
NRFHR = non-reassuring fetal heart rate; VE = vacuum extraction
|
|
Epidural (n, %)
|
33
|
16.0%
|
639
|
20.2%
|
0.150
|
166
|
18.8%
|
0.356
|
301
|
20.7%
|
0.114
|
172
|
20.6%
|
0.141
|
|
Intrapartum fever (n, %)
|
10
|
4.9%
|
288
|
9.1%
|
0.038
|
50
|
5.7%
|
0.647
|
129
|
8.9%
|
0.051
|
109
|
13.0%
|
0.001
|
|
Vacuum cup type
|
Silastic
|
42
|
22.1%
|
1120
|
37.3%
|
< 0.001
|
279
|
33.5%
|
0.002
|
513
|
37.1%
|
< 0.001
|
328
|
41.6%
|
< 0.001
|
|
Kiwi
|
148
|
77.9%
|
1883
|
62.7%
|
|
553
|
66.5%
|
|
869
|
62.9%
|
|
461
|
58.4%
|
|
|
Procedure duration (min) mean ± SD
|
4.5 ± 3.2
|
5.2 ± 3.7
|
0.012
|
4.8 ± 4.3
|
0.359
|
5.3 ± 3.4
|
0.003
|
5.5 ± 3.3
|
0.001
|
|
Vacuum detachment (n, %)
|
41
|
21.4%
|
610
|
20.4%
|
0.763
|
155
|
18.7%
|
0.395
|
282
|
20.5%
|
0.794
|
173
|
22.2%
|
0.805
|
|
Position occiput-anterior (n, %)
|
145
|
70.4%
|
2353
|
74.5%
|
0.190
|
643
|
72.8%
|
0.482
|
1073
|
74.2%
|
0.244
|
637
|
76.8%
|
0.054
|
|
Head station at time of VE (n, %)
|
Mid-pelvis
|
103
|
57.5%
|
1670
|
57.1%
|
0.914
|
452
|
56.0%
|
0.709
|
766
|
57.2%
|
0.932
|
452
|
58.2%
|
0.877
|
|
Low
|
67
|
37.4%
|
1187
|
40.6%
|
0.400
|
329
|
40.8%
|
0.410
|
544
|
40.6%
|
0.413
|
314
|
40.4%
|
0.463
|
|
Outlet
|
9
|
5.0%
|
66
|
2.3%
|
0.019
|
26
|
3.2%
|
0.237
|
29
|
2.2%
|
0.021
|
11
|
1.4%
|
0.002
|
|
Vacuum indication (n, %)
|
NRFHR
|
156
|
85.2%
|
2208
|
74.3%
|
0.001
|
656
|
80.2%
|
0.115
|
1012
|
74.4%
|
0.001
|
540
|
68.2%
|
< 0.001
|
|
Prolonged second stage
|
20
|
10.9%
|
718
|
24.2%
|
< 0.001
|
153
|
18.7%
|
0.019
|
325
|
23.9%
|
< 0.001
|
240
|
30.3%
|
< 0.001
|
|
Maternal indication
|
7
|
3.8%
|
45
|
1.5%
|
0.017
|
9
|
1.1%
|
0.018
|
24
|
1.8%
|
0.062
|
12
|
1.5%
|
0.042
|
|
Labor duration, minutes, mean ± SD
|
First stage
|
432.5 ± 266.0
|
655.0 ± 425.6
|
0.009
|
637.8 ± 376.4
|
0.009
|
685.2 ± 460.1
|
0.007
|
624.1 ± 417.1
|
0.027
|
|
Second stage
|
103.9 ± 79.0
|
140.9 ± 82.0
|
< 0.001
|
131.3 ± 81.9
|
< 0.001
|
141.3 ± 82.4
|
< 0.001
|
150.3 ± 80.2
|
< 0.001
|
|
Third stage
|
10.0 ± 8.0
|
9.5 ± 6.9
|
0.295
|
9.8 ± 7.8
|
0.776
|
9.4 ± 6.6
|
0.270
|
9.1 ± 6.1
|
0.102
|
The duration of VE was significantly shorter in women delivering LBW neonates, and increased with increasing neonatal BW (4.5 minutes in < 2500 g group vs. 4.8, 5.3 and 5.5 minutes in
the 2500–2999 g, 3000–3499 g and ≥ 3500 g groups respectively, p = 0.012).
Fetal head stations at the beginning of extraction were also similar. In all BW groups, most of the fetuses were at S+1 (mid-pelvis) when VE started. Lower percentages were at S+2 (low) and
a minority were at S+3 and below (outlet). However, we found a relatively higher proportion at the outlet station in the LBW group (5% vs. 1.4–3.2%, p = 0.019).
NRFHR was the most common indication for VE, and its rate decreased significantly with every additional 500 g of neonatal BW (85.2% in the LBW group and 80.2%, 74.4% and 68.2% in the
2500–2999 g, 3000–3499 g, ≥ 3500 g groups, respectively, p < 0.001). Prolonged second stage was the second most common indication, and its rate increased significantly with every
additional 500 g of neonatal BW (10.9%, 18.7%, 23.9% and 30.3%, respectively; p < 0.001).
Duration of first and second stages of labor were the shortest for LBW neonates and increased significantly with increasing BW, as presented in [Table 6].
The most common cup type used in all weight groups was the Kiwi OmniCup. It was used most often in the LBW group and decreased with increasing neonatal BW. The Ventouse-Mityvac was used
least for LBW neonates, and its use increased with increasing neonatal birthweight (22.1% among neonates weighing < 2500 g, vs. 33.5%, 37.1% and 41.6% in the 2500–2999 g, 3000–3499 g,
≥ 3500 g groups respectively, p < 0.001).
No differences were noted between groups regarding rates of epidural anesthesia, occiput-anterior (OA) position rates and vacuum cup detachment rates.
Neonatal outcomes of SGA and LGA neonates
Since neonatal BW should be correlated to GA, we also evaluated neonatal outcomes stratified by GA-related BW, comparing SGA, AGA and LGA neonates. Our cohort included 374 (11.1%) SGA
neonates, 2810 (83.2%) AGA neonates and 193 (5.7%) LGA neonates.
Compared to the AGA group, the SGA group had lower rates of SGH (0.5% vs. 4.8%, p < 0.001), lower rates of cephalohematomas (1.1% vs. 3.3%, p = 0.015), and lower rates of total head
injuries related to birth trauma (1.6% vs. 8.3%, p < 0.001). No cases of SDH or intracranial hemorrhage occurred in SGA neonates. When assessing for general neonatal outcomes, no
differences were noted between the groups including NICU hospitalization rates (4.3% vs. 2.8%, p = 0.107). Basic maternal characteristics were similar between groups except for higher rates
of nulliparity (87.2% vs. 77.7%, p < 0.001) and lower rates of gestational diabetes (4.3% vs. 7.7%, p = 0.018) in the SGA group compared to the AGA group. Comparing delivery
characteristics revealed lower rates of epidural anesthesia in the SGA group (15.8% vs. 20.3%, p = 0.038), as well as lower rates of prolonged second stage as the indication for VE (12.0%
vs. 23.8%, p < 0.001).
Compared to the AGA group, the LGA group had higher rates of SGH (13.0% vs. 4.8%, p < 0.001) and higher rates of total head injuries related to birth trauma (15.5% vs. 8.3%, p = 0.001),
although rates of cephalohematoma were similar between groups (2.6% vs. 3.3%, p = 0.570). No cases of SDH or intracranial hemorrhage occurred in LGA neonates. When assessing for general
neonatal outcomes, no differences were noted between the groups except for respiratory distress which was more common for LGA neonates (2.1% vs. 0.6%, p = 0.036). NICU hospitalization rates
were similar between groups (3.1% vs. 2.8%, p = 0.786). Basic maternal characteristics were also similar between groups, except for lower rates of nulliparity in the LGA group compared to
the AGA group (69.4% vs. 77.7%, p = 0.008). Comparing delivery characteristics revealed higher rates of prolonged second stage as the indication for VE in the LGA group (39.1% vs. 23.8%,
p < 0.001). Use of the Ventouse-Mityvac cup was more common in the LGA group (43.6% vs. 36.3%, p = 0.044).
Discussion
Principal findings
We aimed to evaluate the impact of VE on neonatal head injuries related to birth trauma in newborns with a GA ≥ 34 w with LBW (< 2500 g) compared to higher birthweights, using a
non-metal vacuum cup. This comparison allowed the evaluation of VE safety in LBW neonates.
We found that the lowest rates of SGH and cephalohematomas occurred in neonates with BW < 2500 g and increased with every additional 500 g of neonatal weight. Logistic regression found
increasing neonatal weight to be a significant risk factor for neonatal head injuries. When evaluating neonatal outcomes unrelated to birth trauma, no differences were noted between the
groups, except for NICU hospitalization rates, which were the highest among neonates weighing < 2500 g, mainly for observation.
According to our data, the overall rate of SGH in VE was 4.8%. This prevalence is similar to that in previously published studies [23]
[24]. We found lower rates of SGH in neonates with LBW (0.5% vs. 3.5–8.0%, p = 0.001). These findings were found
in some previous studies but not in all. Plauché [25] identified both preterm delivery and fetal macrosomia as risk factors for SGH,
while Aviram et al. [5] found no differences in SGH rates for different BW. Our study was the first large cohort study to look at a
specific population of newborns using non-metal cup VE.
The overall rate of cephalohematomas in VE in our study was 3.1%, one third of the rate reported in the literature [2]. Rates of
cephalohematomas were also lowest among neonates with LBW (0.5% vs. 2.6–3.7%, p = 0.020). These findings are probably related to underdiagnosis but may also be related to the specific cup we
used. These findings have yet to be demonstrated in further studies, as previous studies showed no difference in cephalohematoma rates between different BW groups [5]
[26].
VE is associated with specific risks due to the indications for the procedure and due to the procedure, itself. Nevertheless, we found that neonates in the < 2500 g group had a lower
total rate of head injuries related to VE compared to neonates with higher weights. This result agrees with other studies [5]
[10]
[11] and further expands their findings, since our study is the first to
demonstrate a lower risk of injury due to VE among LBW neonates compared to those with higher birthweights, and not just no difference between groups.
Our findings may be explained by the assumption that lower fetal weight, accompanied by a smaller head diameter, eases the passage of the fetal head through the pelvic canal, thereby making
the VE shorter and less traumatic for the fetus. This explanation agrees with our findings of a higher percentage of outlet stations among the LBW neonates compared to those with higher
BW.
We found NICU hospitalization rates to be highest among LBW neonates. These findings agree with earlier observations in late preterm infants [5]
[27]
[28]. With that said, NICU hospitalization for LBW
neonates is part of our institutional medical protocol and was mainly for observation. This could also explain why NICU hospitalization rates were similar among SGA, AGA and LGA neonates:
the selection of neonates for NICU hospitalization in our institution is based on absolute neonatal birthweight rather than the GA-related BW percentiles.
Prolonged second stage as an indication for VE increased significantly with every additional 500 g of neonatal BW, as did the durations of the first and second stages of labor. This is a
reasonable difference because high fetal weight is a known risk factor for prolonged first and second stages of labor [29]
[30]. Prolonged labor may explain why we found higher rates of intrapartum fever in women delivering higher BW neonates.
VE duration was shorter in women delivering LBW neonates and increased with increasing BW. Logistic regression found increasing procedure duration to be a significant risk factor for
neonatal head injuries. These findings agree with studies showing higher rates of neonatal skull injuries with increasing procedure duration [26]
[31]. After adjusting for confounding variables, increasing BW remained a significant risk factor for neonatal
head injuries. Thus, differences in procedure duration cannot fully explain the lower rates of head injuries we found in the LBW neonates.
The use of the Ventouse-Mityvac cup type was the lowest among the LBW neonates and increased with increasing birthweight. This could be explained by the tendency of the Kiwi-Omni cup to
detach and thus, operators prefer to avoid using it when estimated fetal weight is high. Interestingly, when performing a logistic regression analysis, the Ventouse-Mityvac cup was found to
be an independent risk factor for neonatal head injuries, in addition to increasing neonatal BW and procedure duration. These results might be explained by the tendency of the Kiwi to
detach, making this cup protective, as it prevents the operator from applying too much force to the newborn’s skull compared to the Ventouse-Mityvac. However, as mentioned earlier,
detachment rates were similar between the BW groups. Another explanation could be the lack of flexibility of the Ventouse-Mityvac traction pole, making it more traumatic for the newborn’s
skull. Previous studies evaluating the Ventouse-Mityvac compared to the Kiwi Omnicup only found the metal cup to be associated with neonatal adverse outcomes [16]
[32].
Strengths and limitations
The strengths of our study include the large homogeneous cohort, as data were collected from a single institution with the same protocol for VE, and included only VEs using a non-metal
cup.
This study is not free of limitations. Due to its retrospective design, it lacks important data, such as neonatal head circumference. However, we believe that circumferences matched
neonatal BW and increased proportionally with increasing BW. Factors such as technique and operator experience, as well as the location of the cup in relation to the fetal head sutures were
missing and might have influenced the development of head injuries. Our cohort may have a selection bias, because birthweight was not known at the time of delivery, and most of the small
neonates were probably considered by the senior physician large enough to undergo VE. We chose to include the cases of failed vacuum extraction with a subsequent CD in our cohort in order
not to eliminate neonates with birth trauma caused by an attempt at vacuum extraction from the analysis. We cannot fully exclude the possibility of additional trauma caused by the
manipulation of the fetal head during CD, yet given the low rates of failed VE in our cohort (0.01%) and given the fact that previous reports associate second-stage CD mainly with fractures
and not hematomas [33], we believe the total effect of these cases on our results is negligible.
As far as we know, this is the first cohort comparing VE head injuries related to birth trauma in different neonatal weights and demonstrating a lower risk of head injuries in LBW neonates.
This is also the largest study performed regarding VE and LBW neonates using non-metal vacuum cups. Since prospective randomized studies are unlikely to be performed, we believe our data are
valuable and may useful when consulting with patients.
Conclusion
VE delivery of neonates weighing < 2500 g at 34+ weeks of gestation seems to be a safe mode of delivery when indicated. It resulted in lower rates of head injuries related to birth trauma,
compared to neonates with higher birthweights. This information may be useful in the delivery room, for counseling and reassuring patients, as well as for the medical staff performing VE.
Additional larger-scale studies are needed to establish the outcomes of VE in lower birthweights or for younger GA.
Declarations
Availability of data and material: Data can be made available upon reasonable request from the corresponding author.
Ethics approval: The study was approved by the Meir Medical Center Ethics Committee in August 2019, approval number 0246–19-MMC. Since the study was based on patient records, informed
consent was not required.
Funding: This study was not funded.
Contributorsʼ Statement
Gal Cohen: project development, data collection, manuscript writing and editing. Hanoch Schreiber, Nir Mevorach: data collection. T Biron-Shental, Gil Shechter-Maor, Ofer Markovitch: revised
manuscript critically.
