Keywords
perineal trauma - retrospective cohort study - birth position - obstetric anal sphincter
injuries
Palavras-chave
trauma perineal - estudo retrospectivo de coorte - posição vertical - lesões obstétricas
do esfíncter anal
Introduction
Perineal trauma is an outcome that has received attention in the obstetric and urogynecological
research field. It is prevalent among deliveries,[1] and severe perineal injuries, also called obstetric anal sphincter injuries (OASIs),
are a potential complication of vaginal delivery.[2] Several risk factors are known, and one of the most controversial is the position
of the patient during the second stage of labor.
The lack of a consensus regarding birth position is still evident. A Swedish study
found the lowest rates for OASIs among women giving birth in the standing position,
and the highest rates among women in the lithotomy position.[3] Soong and Barnes[4] divided the birth positions into a variety of types, and the semi-recumbent position
was associated with the need to suture perineal trauma, whereas the all-fours position
was a protective factor to injury in the perineum. Meyvis et al[5] have found that the lateral position resulted in less perineal trauma, and that
the lithotomy position was associated with more episiotomies than other positions.
Moreover, a cohort study of planned home births in 4 Nordic countries found a low
prevalence of OASIs (0.7%) and episiotomy (1%) with most women giving birth in flexible
sacrum positions.[6] Finally, an updated Cochrane review about the position in the second stage of labor
suggested that the upright posture without epidural anesthesia would increase the
risk of second-degree tears and reduce the episiotomy rates.[7]
The need to encourage women to decide in which position they want to give birth is
essential for a humanized approach, especially in a country like Brazil, known for
presenting high rates of cesarean section.[8] Considering this, we have aimed to look for the prevalence of perineal trauma in
a low-risk pregnancy facility that has a multidisciplinary team (midwife and obstetrician)
assisting deliveries, as well as to search for risk factors.
Methods
Study Design, Inclusion/Exclusion Criteria
A retrospective cohort study of 264 singleton pregnancies during labor was performed
at Maternidade Cidinha Bonini, a low-risk pregnancy facility at the Universidade de
Ribeirão Preto (Unaerp, in the Portuguese acronym), Brazil. This maternity cares for
any pregnant woman who has received her prenatal care at any Brazilian Public Unified
Health System (SUS, in the Portuguese acronym) outpatient clinic and is referred to
this hospital for labor. The Institutional Review Board approved the study (under
CAAE 61392616.0.0000.5498). The exclusion criteria were women with preterm birth,
with comorbidities, and patient records in which more than 50% of data were absent.
The study followed the strengthening the reporting of observational studies in epidemiology
(STROBE) checklist for observational studies.
Variables
Perineal trauma (main outcome) was divided into three categories: no tears; first/second-degree
tears with episiotomy; and third and fourth-degree tears. The classification of the
Royal College of Obstetricians and Gynecologists (RCOG) was also used:[9] first and second degrees (mild perineal trauma), third and fourth degrees (severe
perineal trauma). There was no standardized pattern about hands-on or hands-off at
the second stage of labor. The independent variables were: age (divided into < 25
or ≥ 25 years old), race, financial income, parity, marital status, intrapartum analgesia,
birth weight, type of health professional (obstetrician or midwife), and birth position
(lithotomy or upright). The upright or vertical position represents all possibilities
of non-lithotomy, non-supine or non-lateral position (such as kneeling, all-fours,
squatting and standing). Episiotomy was presented in two ways: it was categorized
as a second-degree tear, and as a dummy variable (yes/no).
Statistical Analysis
Data were tabulated in Microsoft Excel for Windows (Microsoft Corporation, Redmond,
WA, USA). The Chi-squared test was utilized for the binomial variables. A multinomial
logistic regression was performed; all variables were inserted simultaneously into
the model, and variable dropout occurred for each variable with the highest p-value. The procedure was repeated so that only statistically significant variables
remained in the final model. The significance level was stipulated at 5%. The statistical
analysis was performed using the Intercooled Stata version 13.0 (StataCorp, College
Station, TX, USA) and R version 3.0.1 (R Foundation for Statistical Computing, Vienna,
Austria) statistical packages.
Results
[Table 1] displays the baseline characteristics of women who underwent labor according to
the presence or absence of perineal trauma. The mean age of the sample was 25.34 ± 5.75
years (range: 13–40), with ∼ 10% of gestations during adolescence, and 68.18% of women
self-reporting as white (). From a total of 264 women, there were 2 cases (0.75%)
of severe perineal trauma, which occurred in nulliparous women younger than 25 years
old. Approximately 46% (121) of the women had no tears, and 7.95% (21) underwent episiotomies
(all of them mediolateral). About the birth position, 42.8% (113) of women preferred
the vertical birth position, while 57.2% (151) preferred the semi-recumbent position,
with no statistical association with perineal trauma (p = 0.285). A total of 76% (200) of the deliveries were performed by obstetricians,
and 23.44% (64) by midwives, with no statistical association with perineal trauma
(p = 0.231). Similarly, newborns weighing 4 kilos or more (p = 0.672), the presence of a sexual partner (p = 0.319), labor analgesia (p = 0.319) and familiar income (p = 0.479) were not associated with perineal trauma.
Table 1
Baseline characteristics of women who underwent vaginal delivery concerning the presence
or absence of perineal trauma
|
Variables
|
Perineal trauma (n/%)
|
p-value
|
|
No tears
|
1st and 2nd degree tear/episiotomy
|
3rd and 4th degree tears
|
|
|
Age
|
|
< 25 years
|
51 (38.35)
|
80 (60.15)
|
2 (1.50)
|
0.019*
|
|
> 25 years
|
70 (53.44)
|
61 (46.56)
|
0 (0)
|
|
Skin color (self-reported)
|
|
White
|
71 (39.44)
|
107 (59.44)
|
2 (1.1)
|
< 0.005*
|
|
Non-white
|
50 (57.5)
|
34 (42.5)
|
0 (0)
|
|
Marital status
|
|
With partner
|
66 (51.56)
|
61 (47.65)
|
1 (0.78)
|
0.088*
|
|
Without partner
|
55 (40.44)
|
80 (58.82)
|
1 (0.73)
|
|
Paid income
|
|
No
|
74 (46.54)
|
83 (52.20)
|
2 (1.26)
|
0.887*
|
|
Yes
|
47 (44.76)
|
58 (55.24)
|
0 (0)
|
|
Parity
|
|
Nulliparous
|
43 (31.85)
|
90 (66.67)
|
2 (1.48)
|
0.005*
|
|
With previous vaginal delivery
|
73 (61.34)
|
46 (38.66)
|
0 (0)
|
|
With previous cesarean
|
5 (50)
|
5 (50)
|
0 (0)
|
|
Birth position
|
|
Lithotomy
|
58 (51.33)
|
54 (47.79)
|
1 (0.88)
|
0.135*
|
|
Upright
|
63 (41.72)
|
87 (57.82)
|
1 (0.66)
|
|
Health professional
|
|
|
|
|
Physician
|
67 (41.88)
|
92 (57.50)
|
1 (0.63)
|
0.190*
|
|
Midwife
|
26 (53.06)
|
22 (44.90)
|
1 (2.04)
|
|
|
Newborn birth weight
|
|
< 3 kg
|
43 (51.19)
|
41 (48.81)
|
0 (0)
|
1.000*
|
|
3–4 kg
|
74 (43.27)
|
95 (55.56)
|
2 (1.17)
|
|
> 4 kg
|
4 (44.44)
|
5 (55.56)
|
0 (0)
|
|
Intrapartum analgesia
|
|
No
|
70 (50)
|
70 (50)
|
0
|
0.173*
|
|
< 6 cm
|
29 (44.62)
|
35 (53.85)
|
1 (1.54)
|
|
> 6 cm
|
22 (37.29)
|
36 (61.02)
|
1 (1.69)
|
Note: *Chi-squared test.
[Table 2] investigates the multinomial regression with two possible dependent variables (episiotomy
and perineal trauma). When the variable perineal trauma was converted to a binomial
fashion (yes/no), there was a statistical association with women younger than 25 years
old (p = 0.019), those who were white (p ≤ 0.005), and nulliparous women (p < 0.005). A multinomial analysis showed that white and nulliparous women were, respectively,
3.89 (range: 1.52–2.96) and 2.89 (range: 1.69–4.95) times more prone to present perineal
tears. When we considered episiotomy as a dependent and dummy variable, nulliparous
women were 4.81 (range: 1.65–14.07) times more prone to undergo episiotomy.
Table 2
Multivariate analysis with two possible dependent variables (episiotomy and perineal
trauma)
|
Variables
|
Adjusted OR (LL-UL)
|
p-value
|
|
White skin color x episiotomy (yes/no)
|
1.35 (0.34–5.33)
|
0.6676
|
|
White skin color x perineal trauma (yes/no)
|
3.89 (1.52–9.96)
|
0.0045
|
|
Non-white skin color x episiotomy (yes/no)
|
1.08 (0.21–5.58)
|
0.9255
|
|
Non-white skin color x perineal trauma (yes/no)
|
2.07 (0.71–6.03)
|
0.1787
|
|
No previous vaginal delivery x episiotomy (yes/no)
|
4.81 (1.65–14.07)
|
0.0041
|
|
No previous vaginal delivery x perineal trauma (yes/no)
|
2.89 (1.69–4.95)
|
0.0001
|
Abbreviations: LL, lower limit; OR, odds ratio; UL, upper limit.
Discussion
The present study has found that in a low-risk maternity with a high prevalence of
upright position in the second stage of labor (42%), no differences were seen regarding
the prevalence of severe perineal tears (one case in the lithotomy group versus one
case in the upright position). Nulliparous women presented a risk factor for presenting
OASIs or to undergo episiotomy, even though there was a low incidence of episiotomy
in our facility. Younger women and white women were also factors that were associated
with perineal trauma; age and birth weight > 4,000 g did not remain significant after
the multinomial analysis. These results are similar to those in the available literature.[1]
[10]
[11]
[12]
Risk factors are well documented in the literature. A retrospective hospital-based
cohort study in Australia found a 5.4% incidence of severe perineal trauma for nulliparous
women versus 1.7% for multipara in 10,408 singleton vaginal deliveries.[10] A prospective observational study in Southeast England found a 6.6% incidence of
OASIs in nulliparous versus 2.7% in multiparous women.[11] Another Brazilian study has found a 2.5% incidence of severe perineal lacerations.
The same study found that operative delivery, primiparity, epidural anesthesia and
higher gestational ages were associated with OASIs.[12] A study performed in another low-risk maternity from the same municipality as our
study (Ribeirão Preto, Brazil) found a 0.9% incidence of severe perineal trauma.[1]
The impact of the maternal position in perineal trauma is still controversial. In
our sample, the upright position was not associated with severe perineal trauma. A
recent pragmatic, multicenter randomized study with 3,093 nulliparous women comparing
upright versus lying down position did not find differences in the prevalence of OASIs.
However, women could modify their birth position during the second stage whenever
they felt like doing so, and this could include some bias in the results.[13] A recent meta-analysis of the effect of upright positions during the second stage
of labor without analgesia did not find an association with third-degree perineal
laceration incidence.[14] It is known that women who give birth in the lithotomy position do not feel that
this position is helpful, which differs from the opinion of the practitioners, and
this may cause a false impression or correlation with negative intra/postpartum outcomes.[15] Moreover, a mixed method study investigating how the maternal birth position could
influence the experience of fathers during childbirth has found that they were more
likely to have a positive experience, or to feel comfortable or powerful.[16]
We do not have regional data about the prevalence of maternal position during the
second stage of labor, but our Western culture has assumed the lithotomic position
as the traditional one, and it is the most taught birth position to obstetricians.[17] We believe that the strengths of the present study are the sample taken from a Brazilian
public hospital, which comprised pregnant women choosing the upright position to give
birth, and the continuous fashion of data collection: no cases were excluded from
this hospital cohort. Nevertheless, there are some limitations in the present study:
its retrospective fashion, the lack of other variables that could be analyzed as risk
factors (second stage duration, fetal head position, head circumference), and the
selection bias of a low-risk maternity, which excludes some other risk factors (such
as maternal obesity) that could be associated with OASIs. Furthermore, performing
a retrospective post-hoc analysis and using our sample (n = 264) with a 13% significance level between birth position and perineal integrity,
we have calculated a study power of 46%, with the minimum required sample of 442 women
to notice a difference (possible type 2 error).
It is essential to prevent OASIs because their impact on subsequent pregnancy outcomes
is high; the risk is increased five-fold in women who had a severe perineal trauma
in their first delivery.[18] To the best of our knowledge, the maternal position during labor does not have a
significant role in preventing OASIs, but it may have positive effects in the childbirth
experience itself. An online survey was mailed to postpartum women, and respondents
who gave birth on the seat had answered that they were more likely to participate
in the decision-making process during labor and to have the opportunity to choose
their preferred birth position.[19] There are several perineal techniques that can be offered to women (such as warm
compresses during the intrapartum period or perineal massage during the antenatal
period) in order to reduce third and fourth-degree tears, and this educational step
may empower these patients.[20] Finally, more prospective studies with different positions than the gynecological,
with women spending most of the labor on their preferred position, avoiding the use
of instrumental deliveries and with larger samples, will be necessary to answer this
question.
Conclusion
The present study has found that the incidence of severe perineal trauma was low,
similar to the incidence in the available literature. The prevalence of upright position
during the second stage of labor was 42%, a high percentage when compared with most
of the birth positions found in epidemiological studies. White and nulliparous women
were more prone to develop perineal tears.
