You are here

OASI: a preventable injury?

European Journal of Obstetrics & Gynecology and Reproductive Biology, pages 9 - 12



The aim of this study was to determine risk factors for obstetric anal sphincter injury and whether any of them were modifiable.

Study design

This was a retrospective review of 2572 women (cases = 1286; controls = 1286) that took place over a 10 year period at a University teaching hospital. Maternal (Age, Parity, BMI and ethnicity), Obstetric (gestational age, assistance during delivery, episiotomy) and fetal (weight) risk factors were analyzed using logistic regression model presented as odds ratio (OR) with 95% confidence intervals (CI). Both univariate and multivariate analyses were conducted with outcome variables comparing cases and controls. Cases without instrumental deliveries were also compared to controls to exclude for the effect of assisted delivery.


This study shows that in addition to instrumental delivery, primiparity (OR 9.8; CI 7.8–12.3), episiotomy (OR 8.6; CI 6.4–11.6), gestational age over 41 weeks (OR 1.5; CI 1.2–1.9), fetal weight over 4 kg (OR 3.2; CI 2.3–4.4) and Asian ethnicity (OR 1.9; CI 1.4–2.7) were all strongly associated with OASI. A raised BMI over 30 appeared to have a protective effect (OR 0.4; CI 0.2–0.5).


Most risk factors related to OASI are non-modifiable however gestational age and episiotomy are modifiable risk factors.

Keywords: OASI, Risk factors, Episiotomy, Instrumental delivery.


OASI (obstetric anal sphincter injuries) which include third (3a, b and c) and fourth degree tears complicate 0.5–1.5% of vaginal deliveries [1] and are associated with significant morbidity. About 30–50% of these women suffer from chronic anal incontinence, dyspareunia, fecal urgency or perineal pain[2] and [3]. The seriousness of the complications after OASI is related to the severity of sphincter injury. There is no consensus regarding preventive measures and clinical management of severe perineal tears. There are also conflicting data regarding the significance of various obstetric risk factors for such tears. Several studies have identified a number of obstetric risk factors associated with sphincter injury. Factors consistently shown to be associated with perineal tears are instrumental delivery, with forceps associated with a higher risk than ventouse, longer duration of second stage of labor, nulliparity, birth weight and occipito posterior position[1], [4], and [5]. However, many of the studies in the literature are contradictory on other factors such as episiotomy and ethnicity [6] . Establishment of risk factors for such tears may enable earlier identification of patients at risk and the use of preventive measures.

The aim of this study was to determine possible risk factors for third and fourth degree perineal tears in a single university-affiliated maternity hospital with approximately 8000 deliveries per year and strategies for prevention.

Materials and methods

This is a retrospective case-control study over a ten-year period from January 1st 2003 to December 31st 2012. The study was performed at the Jessop Wing Hospital, Sheffield Teaching Hospitals a tertiary-referral University hospital where approximately 8000 deliveries take place annually. The incidence of OASI in this unit is 3% but fluctuates between 1 and 5% which is within national averages. The study population comprised 2572 women (cases = 1286; controls = 1286), who met the following inclusion criteria: singleton pregnancy, vertex presentation, and vaginal delivery. Multifetal pregnancies, breech presentations, and caesarean sections were excluded from the analysis. Case records of all women who sustained an OASI were identified from the maternity record database and examined for the presence of risk factors. Women without anal sphincter injury, matched for maternal age with each case was identified from the maternity database as a control. The control group did not have an instrumental delivery as this is a known risk factor for OASI [3] .

OASI was assessed and classified clinically using the accepted Standardised classification system of the Royal College of Obstetricians and Gynaecologists (RCOG). Though, anal endosonography is more accurate and provides more information on the extent of OASI compared to clinical and conventional physiological methods [7] , it is not routine practice to perform scans to confirm diagnosis and current RCOG recommendations are that this should be a clinical diagnosis (RCOG, Green Top Guideline no 29; 2007). In this study, no distinction was made between the third and fourth degree perineal tears for the analysis of data.

The following data on maternal characteristics were collected: age, parity, body mass index (BMI) and ethnicity. Ethnicity was categorized as White Caucasians, Asian, Black and mixed origins. Parity was classified into two categories: women delivering their first child (“primiparous”) and women with a history of one or more vaginal deliveries (“multiparous”). BMI was divided into 3 categories: <30, 30–35 and >35).

Obstetric data collected were: induction of labor, episiotomy performed, gestational age at the time of delivery >41 weeks which had an affirmative or negative response (yes/no) and the type of instrumental deliveries (forceps or vacuum).

The infant variables identified were gestational age at the time of delivery and birth weight (kg).

The relationship between potential maternal, obstetric and infant related risk factors for OASI were analyzed using logistic regression model presented as odds ratio (OR) with 95% confidence intervals (CI).

This study was conducted as a service evaluation project so formal ethical approval was not required.

Both univariate and multivariate analyses were conducted with the outcome variables comparing cases and controls. Women who sustained an OASI without an instrumental delivery were also compared to controls by conducting multivariate analysis to exclude for the effect of instrumental delivery. These analyses were done, using SPSS version 21. The level of significance was set atP < 0.05.


One thousand two hundred and eighty six women with OASI and the same number of age matched controls were identified during the study period. The controls were of similar age (29.5 ± 5.8 vs 28.5 ± 5.6), BMI (25.9 ± 19.3 vs 26.6 ± 6.5) and ethnicity.

Descriptive statistics for the subjects and birth are shown in Table 1 . Cases and control subjects differed significantly on most predictor variables. Cases were more likely to be nulliparous (59.6% vs 11%), delivered heavier infants at mean gestational age 39.9 weeks and almost half of them (44.3%) had instrumental deliveries. In addition, deliveries of patients with OASI was more likely to involve an episiotomy (58.2% vs 5.1%) than in the control group. The Univariate analysis identified several factors as being associated with a higher risk of OASI ( Table 2 ). The risk of OASI was 12 times higher in primiparous (OR 11.9; 95% CI 9.7–14.6] compared to multiparas. Strong associations were observed for obstetric intervention such as an episiotomy (OR 13.5; 95% CI 10.3–17.8). Larger babies (OR 2.6; 95% CI 2.0–3.3) and advanced gestational age (OR 1.9; 95% CI 1.6–2.2] at the time of delivery were also associated with a higher risk of OASI. However, increased maternal age (>40 years) and BMI (>30) seemed to be 60% protective against OASI.

Table 1 Demographics of women, labor and birth with and without OASI.

Variables Cases n = 1286 Control n = 1286 P value
Age (yrs) 28.5 ± [5.6] 29.5 ± [5.8] 0.12
BMI (kg/m2) 25.9 ± [19.3] 26.6 ± [6.5] 0.23
Primiparity 767 (59.6) 142 (11.0) <0.0001
Ethnicity 965 (75.0) 976 (75.9) 0.20
 White 126 (9.8) 111 (8.6)
 Asian 38 (3.0) 55 (4.3)
 Black 157 (12.2) 144 (11.2)
Induction of labor 320 (24.9) 284 (22.1) 0.25
Gestational age (week) 39.9 ± [1.3] 39.1 ± [2.4] <0.0001
Method of delivery     <0.0001
 Spontaneous 716 (55.7) 1286 (100)
 Forceps delivery 394 (30.6) 0 (0)
 Vacuum delivery 176 (13.7) 0 (0)
Episiotomy performed 538 (58.2) 65 (5.1) <0.0001
Infant birth weight (kg) 3.6 ± [0.5] 3.2 ± [0.6] <0.0001

Table 2 Univariate analysis for risk factors in women with and without OASI.

Characteristics Odds ratio [95% confidence interval] Overall P value
Age (reference = 20–40)   <0.0001
 <20 yrs 2.8 [1.8–4.2]
 >40 yrs 0.4 [0.2–0.7]
Parity (reference = multiparous)   <0.0001
 Primiparous 11.9 [9.7–14.6]
Ethnicity (reference = white)   =0.20
 Asian 1.2 [0.8–1.5]
 Black 0.7 [0.5–1.1]
 Mixed 1.1 [0.9–1.4]
BMI (kg/m2)   <0.0001
 Reference ≤ 30  
 Between 30 and 35 0.5 [0.4–0.7]
 >35 0.4 [0.2–0.5]
Gestational age (reference ≤ 41 weeks)   <0.0001
 >41 weeks 1.9 [1.6–2.2]
Labor (reference =spontaneous)   =0.25
 Induction 0.1 [0.06–15.4]
Episiotomy (reference = no)   <0.0001
 Yes 13.5 [10.3–17.8]
Birth weight   <0.0001
 Reference ≤ 4 kg  
 >4 kg 2.6 [2.0–3.3]

A multivariate analysis was then done for all the variables ( Table 3 ). The largest independent protective effect was multiparity, with a 10 times increased risk of OASI in primipara (OR 9.8; 95% CI 7.8–12.3) compared to multiparous. Ethnicity, which was not found to have an effect on OASI in Univariate analysis, was observed to be a significant independent risk factors (P = 0.001) for third and fourth degree perineal tears in the multivariate logistic regression analysis. Women of Asian ethnic origin were 2 times at higher risk of OASI (OR 1.9; 95% CI 1.4–2.7) compared to white British born women. Higher gestational age at the time of delivery, heavier babies and increased rate of episiotomy were strong risk factors for OASI (P < 0.0001). Increased maternal age no longer remained protective in the multivariate model (P = 0.08). However, raised BMI was still found to be protective (OR 0.4; 95% CI 0.2–0.5) for the occurrence of OASI.

Table 3 Significant independent risk factors in multivariate logistic regression model..

Characteristics Odds ratio [95% confidence interval] Overall P value
Age   =0.08
 <20 yrs 1.7 [1.1–2.9]
 >40 yrs 0.7 [0.4–1.5]
Ethnicity   =0.001
 Asian ethnicity 1.9 [1.4–2.7]
Parity   <0.0001
 Primiparity 9.8 [7.8–12.3]
BMI   =0.003
 30–35 0.7 [0.5–1.0]
 >35 0.5 [0.3–0.8]
Gestational age   <0.0001
 >41 weeks 1.5 [1.2–1.9]
Labor   =1.0
 Induction 1.0 [0.04–26.4]
Episiotomy 8.6 [6.4–11.6] <0.0001
Birth weight>4 kg 3.2 [2.3–4.4] <0.0001

Because our control group did not include any patients with instrumental delivery, a further sub-analysis was performed on cases who suffered an OASI, following spontaneous vaginal delivery (excluding the instrumental deliveries) (n = 716) and compared them with the controls (n = 1286) ( Table 4 ). Factors such as primiparity, Asian ethnicity, higher gestational age, heavier babies, and episiotomy which were found to have a significant effect on OASI in the multivariate logistic regression analysis in Table 3 , remained statistically significant independent risk factors, after excluding 44.3% (n = 570) cases with instrumental deliveries.

Table 4 Comparison of risk factors for OASI in controls (n = 1286) and cases (716) without instrumental vaginal delivery.

Characteristics Women without OASI (control) n = 1286 Women with OASI (cases) n = 716 Odds ratio [95% confidence interval] Overall P value
Ethnicity       <0.0001
 Asian ethnicity 111 (8.6) 76 (10.6) 2.2 [1.5–3.1]
Parity       <0.0001
 Primiparous 142 (11.0) 385 (53.8) 9.1 [7.1–11.7]
BMI       =0.02
 30–35 193 (15) 70 (9.8) 0.8 [0.5–1.1]
 >35 108 (8.4) 33 (4.6) 0.6 [0.4–0.9]
Gestational age       =0.009
 >41 weeks 286 (22.2) 217 (20.1) 1.4 [1.1–1.8]
Episiotomy       <0.0001
 Yes 65 (5.1) 144 (20.1) 2.4 [1.7–3.4]
Birth weight       <0.0001
 >4 kg 95 (7.4) 133 (18.6) 3.7 [2.7–5.2]


This study shows that primiparity, instrumental delivery, episiotomy, gestational age over 41, gestational weight over 4 kg and Asian ethnicity were all strongly associated with OASI. A raised BMI over 30 appeared to have a protective effect.

The strength of this study is the size and heterogeneity of the population studied. This study also identifies increasing gestational age as an independent risk factor, not previously seen in other studies. This is one of the few isolated modifiable risk factors. In this study we also report on risk factors in cases with OASI after excluding instrumental deliveries which has not been analyzed in other studies.

We acknowledge that case-control studies like this one are prone to bias resulting in both over and under-estimation of the true risk, hence findings from such studies should be interpreted with a degree of caution. Inherent limitations in a retrospective analysis include the inability to control data quality, because all the data are from medical records and the potential inconsistency in record keeping across time and across clinicians. However, this means that data were entered without knowledge of a study hypothesis, therefore reducing the risk of bias.

We were unable to control for a number of risk factors that might have influenced the results. For example the type of anesthesia used[1] and [3]the experience and training of the birth attendant [8] , slowing the delivery of the infant's head and instructing the mother not to push while the head is delivered [9] , position of the head in the pelvis [10] , and other obstetric interventions such as maternal position at the time of delivery [4] , perineal protection techniques [11] and the angle and size of an episiotomy at the time of delivery [12] . Furthermore, information on indication for induction of labor such as diabetes, macrodome which could be a potential confounding factor for OASI was not available in our data set.

Mode of delivery is a key determinant of the risk for perineal tears, with studies consistently demonstrating that women with instrumental deliveries have 1.5–14.0 times higher rates of anal sphincter tears compared to spontaneous vaginal delivery[4], [5], and [8]. In accordance with previous reports, our results suggest that the dominant risk factors for anal sphincter injury is primiparity. This may be due to relative inelasticity of the perineum, and a reduction in risk with increasing parity[1], [2], [5], and [8].

Very few studies have considered maternal age as a risk factor for the occurrence of OASI. However, results from the studies which have taken this factor into account are inconsistent. Some studies have shown that due to in elastic perineal tissue, older women are at greater risk of anal sphincter tears [2] , while others have found no such association [13] . We observed increased maternal age to be protective of perineal lacerations in our univariate analysis. However, our multivariate corrected model revealed no such protection. This suggest that there were possible confounders to the maternal age in the Univariate model, which were controlled in the multivariate regression model.

High prevalence of third and fourth degree perineal tears among Asian women delivering in Western countries has been reported by others, and is thought to be associated with a relatively short perineum that is less likely to stretch well, or relatively higher birth weights secondary to dietary changes [14] . Our results are in agreement with the previous studies[5], [8], [13], and [15].

The role of episiotomy in anal sphincter injury has remained controversial[6] and [16]. Several studies cited midline episiotomy as a risk factor that is associated with third and fourth degree tear [4] , although others have not found such association [8] . Small differences in the rates of anal sphincter tears were observed in a randomized controlled trial of routine vs restrictive use of episiotomy at operative vaginal delivery (8.1% vs 10.9%, OR 0.72, 95% CI 0.28–1.87) [16] . All episiotomies given in our study involved a mediolateral incision, which is recommended practice in the UK [6] . We found that episiotomy was associated with an increased risk for OASI both in cases with and without instrumental deliveries. This is in contrast to the finding of Ekéus et al. [15] who found that mediolateral episiotomy was related to an increased risk of anal lacerations in spontaneous deliveries but protective in instrumental deliveries.

We observed a significant positive correlation between birth weight and the occurrence of third and fourth degree perineal tears. Our findings confirm the results of previous studies that birth weight greater than 4 kg is associated with an elevated risk of OASI[8] and [13]. This association is presumably due to the mechanical stress of delivering a large baby.

An interesting finding that emerged from our study was the association of higher gestational age with the occurrence of OASI. This is in agreement with Fenner et al. [17] but contrary to the finding of Samarasekera et al. [18] who found no association of gestational age with the occurrence of anal sphincter injury. Gestational age >41 weeks is considered to be a significant predictor (P < 0.001) for macrosomia [19] . Our data suggest that gestational age > 41 weeks and babies weighing >4 kg acquire third and fourth degree perineal tears 1.5 and 3 times more likely compared to the controls. Multivariate analysis showed that Gestational age was an independent risk factor for OASI irrespective of birth weight.

Surprisingly, results from our study demonstrated significant protection for the occurrence of OASI with increased maternal BMI. Results from our study illustrates 30% and 50% protection in patients with BMI 30–35 and >35, respectively. This is hard to believe as there is strong evidence in literature that confirms that an increased BMI increases the incidence of macrosomia, the need for obstetric intervention and consequent morbidity in mother and baby [20] . The question here is whether the observed reduced risk of anal sphincter injury among obese women is true or false? False in the sense that the decreased risk was due to lower detection rate of injuries in the obese group of women. It could be possible that that the excessive amount of fat tissue in the perineal region complicates adequate examination of the anatomy. However, if the negative association between maternal obesity and risk of anal sphincter injuries is true, a clinical speculation could be that the increased amount of adipose tissue softens the tissue and make it more stretchable, hence protecting against the development of perineal lacerations. This hypothesis is supported by studies who have shown lower risk of perineal laceration with higher BMI[21] and [22].

In conclusion, the majority of risk factors for OASI, such as parity and birth weight, are not modifiable. However, episiotomy continues to be a major potentially modifiable risk factor and efforts should be made to limit this procedure to only when medically necessary. Another modifiable risk factor which was identified from our study was gestational age >41 weeks. Although induction of labor from our study reduced the risk of OASI by 10%, this result was not statistically significant. Prior studies have not supported routine use for induction for the indication of macrosomia on the grounds that it might increase the risk of caesarean section, rather than considering other maternal morbidities such as third and fourth degree perineal tears, leading to long-term anorectal complaints and its distressing and disabling consequence. There is growing body of evidence that induction of labor at term in other high-risk scenarios does not increase emergency caesarean section rates, rather the reverse[23], [24], and [25]. Future research is needed to explore a relationship between advanced gestational age and macrosomia and to compare induction of labor vs expectant management in both groups. Furthermore we need to look at risk modeling for individual women based on modifiable and non-modifiable risk factors so that women can be informed of their risks of suffering an OASI. This will inform decision making and allow women to make choices about their mode of delivery.


Most risk factors related to OASI are non-modifiable (parity, age, instrumental delivery, ethnicity, birth weight) however gestational age and episiotomy may be modifiable risk factors.


  • [1] A. Williams, D.G. Tincello, S. White, E.J. Adams, Z. Alfirevic, D.H. Richmond. Risk scoring system for prediction of obstetric anal sphincter injury. BJOG. 2005;112(8):1066-1069 Crossref
  • [2] V. Revicky, D. Nirmal, S. Mukhopadhyay, E.P. Morris, J.J. Nieto. Could a mediolateral episiotomy prevent obstetric anal sphincter injury?. Eur J Obstet Gynecol Reprod Biol. 2010;150(2):142-146 Crossref
  • [3] C. Dahl, P. Kjølhede. Obstetric anal sphincter rupture in older primiparous women: a case-control study. Acta Obstet Gynecol Scand. 2006;85(10):1252-1258 Crossref
  • [4] E. Eason, M. Labrecque, G. Wells, P. Feldman. Preventing perineal trauma during childbirth: a systematic review. Obstet Gynecol. 2000;95(3):464-471 Crossref
  • [5] A. Groutz, J. Hasson, A. Wengier, R. Gold, A. Skornick-Rapaport, J.B. Lessing, et al. Third- and fourth-degree perineal tears: prevalence and risk factors in the third millennium. Am J Obstet Gynecol. 2011;204(4):347.e1-347.e4 Crossref
  • [6] L.A. Smith, N. Price, V. Simonite, E.E. Burns. Incidence of and risk factors for perineal trauma: a prospective observational study. BMC Pregnancy Childbirth. 2013;13:59
  • [7] V. Andrews, A.H. Sultan, R. Thakar, P.W. Jones. Occult anal sphincter injuries—myth or reality?. BJOG. 2006;113(2):195-200 Crossref
  • [8] V.L. Handa, B.H. Danielsen, W.M. Gilbert. Obstetric anal sphincter lacerations. Obstet Gynecol. 2001;98(2):225-230 Crossref
  • [9] K. Laine, T. Pirhonen, R. Rolland, J. Pirhonen. Decreasing the incidence of anal sphincter tears during delivery. Obstet Gynecol. 2008;111(5):1053-1057 Crossref
  • [10] O. Eskandar, D. Shet. Risk factors for 3rd and 4th degree perineal tear. J Obstet Gynaecol. 2009;29(2):119-122 Crossref
  • [11] M. Eogan, L. Daly, C. O’Herlihy. The effect of regular antenatal perineal massage on postnatal pain and anal sphincter injury: a prospective observational study. J Matern Fetal Neonatal Med. 2006;19(4):225-229 Crossref
  • [12] M. Eogan, L. Daly, P.R. O'Connell, C. O'Herlihy. Does the angle of episiotomy affect the incidence of anal sphincter injury?. BJOG. 2006;113(2):190-194 Crossref
  • [13] I. Gurol-Urganci, D.A. Cromwell, L.C. Edozien, T.A. Mahmood, E.J. Adams, D.H. Richmond, et al. Third- and fourth-degree perineal tears among primiparous women in England between 2000 and 2012: time trends and risk factors. BJOG. 2013;120(12):1516-1525
  • [14] L.M. Hopkins, A.B. Caughey, D.V. Glidden, R.K. Jr. Laros. Racial/ethnic differences in perineal: vaginal and cervical lacerations. Am J Obstet Gynecol. 2005;193(2):455-459 Crossref
  • [15] C. Ekéus, E. Nilsson, K. Gottvall. Increasing incidence of anal sphincter tears among primiparas in Sweden: a population-based register study. Acta Obstet Gynecol Scand. 2008;87(5):564-573
  • [16] D.J. Murphy, M. Macleod, R. Bahl, K. Goyder, L. Howarth, B. Strachan. A randomised controlled trial of routine versus restrictive use of episiotomy at operative vaginal delivery: a multicentre pilot study. BJOG. 2008;115(13):1695-1702 (Discussion 1702–3)
  • [17] D.E. Fenner, B. Genberg, P. Brahma, L. Marek, J.O. DeLancey. Fecal and urinary incontinence after vaginal delivery with anal sphincter disruption in an obstetrics unit in the United States. Am J Obstet Gynecol. 2003;189(6):1543-1549 (Discussion 1549–50) Crossref
  • [18] D.N. Samarasekera, M.T. Bekhit, J.P. Preston, C.T. Speakman. Risk factors for anal sphincter disruption during child birth. Langenbecks Arch Surg. 2009;394(3):535-538 Crossref
  • [19] N.E. Stotland, A.B. Caughey, E.M. Breed, G.J. Escobar. Risk factors and obstetric complications associated with macrosomia. Int J Gynaecol Obstet. 2004;87(3):220-226 Crossref
  • [20] R. Scott-Pillai, D. Spence, C.R. Cardwell, A. Hunter, V.A. Holmes. The impact of body mass index on maternal and neonatal outcomes: a retrospective study in a UK obstetric population, 2004–2011. BJOG. 2013;120(8):932-939 Crossref
  • [21] M. Blomberg. Maternal body mass index and risk of obstetric anal sphincter injury. Biomed Res Int. 2014;2014:395803
  • [22] E.S. Lindholm, D. Altman. Risk of obstetric anal sphincter lacerations among obese women. BJOG. 2013;120(9):1110-1115 Crossref
  • [23] K.F. Walker, G. Bugg, M. Macpherson, C. McCormick, C. Wildsmith, G. Smith, et al. Induction of labour versus expectant management for nulliparous women over 35 years of age: a multi-centre prospective, randomised controlled trial. BMC Pregnancy Childbirth. 2012;12:145 Crossref
  • [24] C.M. Koopmans, D. Bijlenga, H. Groen, S.M. Vijgen, J.G. Aarnoudse, D.J. Bekedam, et al. Induction of labour versus expectant monitoring for gestational hypertension or mild pre-eclampsia after 36 weeks’ gestation (HYPITAT): a multicentre, open-label randomised controlled trial. Lancet. 2009;374(9694):979-988 Crossref
  • [25] M.E. Hannah, A. Ohlsson, D. Farine, S.A. Hewson, E.D. Hodnett, T.L. Myhr, et al. Induction of labor compared with expectant management for prelabor rupture of the membranes at term. TERMPROM Study Group. N Engl J Med. 1996;334(16):1005-1010 Crossref


Department of Urogynaecology, Level 4 Jessop Wing, Sheffield Teaching Hospitals, NHS Trust, Tree Root Walk, Sheffield S10 2SF, United Kingdom

lowast Corresponding author. Tel.: +0044 0 114 2268166; fax: +0044 0 121 6272102.