Dr H. Rydhstroem, Department of Obstetrics and Gynaecology, Helsingborg Hospital, SE-251 87 Helsingborg, Sweden.
Background Injury to the genital tract sustained during childbirth can lead to transient or protracted morbidity. Attention should be paid to avoidable risk factors that can cause this complication.
Aim To analyse the recurrence, at a later delivery, of trauma to the genital tract, subsequent to perineal laceration of the sphincter ani (third or fourth degree), sustained at an earlier delivery.
Design A population-based study.
Setting In Sweden, 1973–1997 inclusive.
Population All women with a vaginal, singleton delivery in Sweden.
Methods The Medical Birth Registry, the National Board of Health and Welfare was used to identify cases of ruptured sphincter ani.
Main outcome measures OR was calculated with 95% confidence interval. A stratified analysis was performed using the Mantel-Haenszel technique.
Major end point Rupture of the sphincter ani (third or fourth degree) at second delivery.
Results The incidence of anal sphincter rupture increased sixfold during the study period, from 0.5% in 1973 to 3.0% in 1997. Women who had sustained a laceration of this type ran a significantly increased risk of a recurrence at a later delivery. This effect persisted even after stratification for birthweight, year of birth, parity and maternal age (OR 4.74, 95% confidence interval 4.34–5.17). When only fourth degree rupture was considered (rupture of both anal sphincter and rectum), the corresponding figures were 6.52 (95% CI 5.29–8.04). This effect also persisted after stratification for birthweight, year of birth, parity and maternal age. The OR for giving birth a second time, subsequent to a third or fourth degree perineal laceration at first delivery, was 0.68 (95% CI 0.67–0.70).
Conclusion Our findings suggest that the risk of an anal sphincter rupture at delivery increases five to sevenfold when there has been a similar rupture at a previous delivery. Further study is needed before safe recommendations can be made concerning the subsequent mode of delivery to be adopted, following rupture in the sphincter ani at a previous birth.
Trauma to the genital tract can occur at birth and cause transient or protracted morbidity.1,2 Earlier studies have described risk factors for third and fourth degree perineal lacerations, including nulliparity, breech presentation, malpresentation of the fetal head, fetal macrosomia, forceps or vacuum delivery, and midline episiotomy.3–6
Because sphincter laceration during vaginal birth is a serious complication and is frequently associated with subsequent anal incontinence, particular attention should be paid to the identification of avoidable risk factors for this complication. It has been suggested that a previous third degree or fourth degree perineal rupture significantly increases the risk of laceration at the next delivery.2 However, most earlier studies have not been population-based, have included relatively few cases and accordingly have had little external validity.
The aim of this population-based study was to determine the recurrence rate at the second birth for women with perineal rupture (third or fourth degree) of the anal sphincter at the first delivery.
Data on parity, maternal age, birthweight and rupture of the sphincter ani were collected from the Medical Birth Registry (MBR), the National Board of Health and Welfare, Stockholm. All singleton pregnancies leading to a vaginal birth between 1st January 1973 and 31st December 1997 were included. Multiple pregnancies were excluded. The first as well as the second birth took place during the study period. The MBR keeps records of all deliveries (including stillbirths) following a gestation lasting at least 28 weeks. Compared with the official statistics, 1–2% of all births are missing from the MBR.7,8
All women with a singleton delivery resulting in a diagnosed and treated laceration of the anal sphincter were identified by the code 658.1 (ICD8), 664.2 (ICD9) or O70.2 (ICD10). A third degree tear was defined as a rupture (partial or complete) through the sphincter but leaving the rectal mucosa intact. A fourth degree tear was identified (and included) by the code 658.2 (ICD8), 664.3 (ICD9) or O70.3 (ICD10) when the rectal mucosa was involved.
Odds ratio (OR) was calculated with 95% confidence intervals according to Miettinen.9 To control for confounding factors such as parity, maternal age, delivery unit and year of delivery, a stratified analysis was performed using the Mantel–Haenszel technique.10
The incidence of such ruptures in the total population (third or fourth degree) increased sixfold during the study period, from 0.5% in 1973 to 3.0% in 1997 (Fig. 1). When including only the first and second vaginal births and excluding twin pregnancies, 17,594 cases of ruptured sphincter ani remained for analysis (Table 1). Maternal age between 25 and 39 years carried an increased risk of rupture of the sphincter (Table 2). Over 39 years of age, an OR above unity was seen, although not a significant increase. Primiparas ran a more than 10-fold greater risk of sustaining a serious laceration compared with parous women (Table 3). Birthweight is a strong predictor of perineal laceration (Table 4). A birthweight above 5 kg increases the risk some 15- to 25-fold compared with a birthweight below 3 kg. In the second delivery, the overall caesarean section rate was 8.8%. In retrospect, there is no possibility of finding out how many of these sections were performed to prevent recurrence of rupture of the sphincter ani.
Table 1. Number of deliveries included in the analysis.
Deliveries first and second delivery combined (n)
Third and fourth degree ruptures (n)
Only fourth degree rupture (n)
Total number of deliveries
Table 2. OR and 95% confidence interval for perineal laceration related to maternal age, after stratification for parity, birthweight, year of delivery and instrumental delivery. All deliveries in Sweden between 1973 and 1997.
Maternal age (years)
Table 3. OR and 95% confidence interval for perineal laceration related to parity, after stratification for maternal age, birthweight, year of delivery, and instrumental delivery. All deliveries in Sweden between 1973 and 1997.
Table 4. OR with 95% confidence interval for perineal laceration related to birthweight, after stratification for parity, maternal age, year of delivery and instrumental delivery. All deliveries in Sweden between 1973 and 1997.
Women with a history of sphincter rupture are at significantly greater risk of sustaining another such injury at the next delivery, irrespective of birthweight or maternal age (Table 5). The overall risk in this group now increased to 4.4. Twenty-three women with a history of previous third or fourth degree injury to the sphincter ani would have to be delivered by caesarean section to avoid one recurrent sphincter ani rupture.
Table 5. Recurrence rate for rupture of sphincter ani in Sweden between 1973 and 1997. OR with 95% confidence interval.
Third and fourth degree rupture
No stratification for birthweight, maternal age, year or parity
Stratification for birth weight, maternal age, year and parity
Only fourth degree rupture
No stratification for birth weight, maternal age, year or parity
Stratification for birthweight, maternal age, year and parity
Inclusion of only fourth degree sphincter rupture at the first delivery did not increase the OR markedly (from 4.74 to 6.52). The OR for having a second child, after a third or fourth degree perineal laceration at the first delivery, was 0.68 (95% CI 0.67–0.70), that is, only 68% (of the expected 100%, as seen for women with no rupture) had another child after the first one.
The results of the present study confirm a significantly increased recurrence rate for mothers who sustained a sphincter ani rupture at first delivery. This is consistent with most previous reports. In a hospital-based study based of 4000 births, Peleg et al.2 found a twofold increased risk at subsequent delivery. Payne et al.6 found a similar association, based on 1741 women who gave birth vaginally between 1990 and 1994. However, the paucity of cases in earlier studies makes direct comparisons with our findings difficult. Our results reflect the situation in Sweden over the past 25 years. The caesarean section rate remained stable at around 10–12% from 1980 to 1996.11 During that period, very few caesarean sections were performed for the indication ‘sphincter rupture at a previous delivery’.
In view of the high recurrence rate, it has been suggested that elective caesarean section would be preferable for the second delivery. It is not known whether women with recurrent anal sphincter rupture experience more severe short and long term problems than women who undergo elective caesarean section. To our knowledge, no such study has been published. However, the ability of the anal sphincter to maintain continence weakens with each subsequent delivery.12,13 This may argue for a more liberal attitude to caesarean section in subsequent pregnancies, or at least that an expectant mother should be given information about the risks of a future (vaginal) delivery, once she has had a repair of a ruptured sphincter ani. It is also important to emphasise that the great majority will not sustain a rupture at their second vaginal delivery. In fact, the ‘numbers needed to treat’ was 23, that is, 23 caesarean sections have to be performed to prevent one recurrent rupture of the sphincter ani and, as is pointed out in the latest analysis of maternal deaths in Britain, not even a planned caesarean section is risk-free.14
The increase in the incidence of anal sphincter rupture in Sweden from 1973 to 1997 is difficult to explain. It is probably partly the result of more accurate registration at the MBR.11 Another explanation may be that in some women a minor defect in the sphincter ani muscle was not diagnosed and accordingly overlooked early in the study period. Without further proof, however, we believe that part of the increase is real.
Interestingly, a previous study15 indicated that the Ritgen Manoeuvre used during delivery of the fetal head may protect the perineum (including the anal sphincter). This manoeuvre encourages early extension of the fetal head. The midwife presses the baby's head with her left hand to control the speed of crowning and simultaneously, the midwife uses the thumb and index finger of the right hand to support the perineum, while the flexed middle finger is used to take a grip on the baby's chin. The woman is asked to stop pushing and to breathe rapidly, while the midwife slowly helps baby's head through the vaginal introitus. When most of the head is out, the perineal ring is pushed under the baby's chin. At one large Swedish hospital, a high rate of sphincter rupture was seen concomitant with a much less frequent use of this manoeuvre, whereas in a large Finnish hospital the situation was the opposite. However, one can only speculate as to whether the use of this thorough manoeuvre decreased in Sweden during the study period, which would explain the increased incidence of rupture of the anal sphincter.
The birthweight has increased in Sweden during recent decades, but not to an extent that would explain the increased incidence of ruptures.16 In that publication, based on the same material as the present analysis, it is stated that birthweight had increased by about 60 g during the last 22 years. For a birthweight over 4.0 kg, the percentage increased from 16.9% during 1973–1977 to 20.3% during 1991–1997.
An interesting but worrying finding in the present study was the significantly reduced frequency of a second child for those women with a ruptured anal sphincter at their first delivery. It is possible that the choice of future elective caesarean section for such women might increase the subsequent pregnancy rates. It is known that a traumatic birth experience has an impact on future reproduction.17
In conclusion, the results of the present study appear to indicate that the risk of a repeat anal sphincter rupture following a previous one increases four- to sixfold. Further studies are needed if safe recommendations are to be given regarding the mode of delivery at a second birth. If future studies can verify a further burden for the mother threatening a second rupture of the sphincter, an abdominal delivery should be recommended.