OBJECTIVE: To investigate risk factors for obstetric anal sphincter injuries in a large population-based data set, and to assess to what extent changes in these risk factors could account for trends in obstetric anal sphincter injuries.
METHODS: This is a population-based cohort study on data from the Medical Birth Registry of Norway between 1967 and 2004, including all vaginal singleton deliveries of vertex-presenting fetuses weighing 500 g or more. Women with their first birth before 1967 and births with previous obstetric anal sphincter injuries were excluded, leaving 1,673,442 births for study. The outcome variable was third- and fourth-degree obstetric anal sphincter injuries. The associations of obstetric anal sphincter injuries with possible risk factors were estimated by odds ratios (ORs) obtained by logistic regression.
RESULTS: The occurrence of obstetric anal sphincter injuries increased from 0.5% in 1967 to 4.1% in 2004. After adjusting for demographic and other risk factors, as well as possible confounders, the increase of obstetric anal sphincter injuries persisted, although reduced (unadjusted OR 7.1; 95% confidence interval [CI] 6.8–7.4; adjusted OR 5.6; 95% CI 5.3–5.9). Obstetric anal sphincter injuries were significantly associated with maternal age 30 years or older, vaginal birth order of one, previous cesarean delivery, instrumental delivery, episiotomy, type 1 diabetes, gestational diabetes, induction of labor by prostaglandin, size of maternity unit, birth weight 3,500 g or more, head circumference 35 cm or more, and African or Asian country of birth.
CONCLUSION: Risk of obstetric anal sphincter injuries considerably increased in Norway in 1967 to 2004. Changes in the risk factors studied could only partially explain this increase.
LEVEL OF EVIDENCE: II
Changes in the risk factors studied only partially explain the increasing occurrence of obstetric anal sphincter injuries in Norway between 1967 and 2004.
From the Institute of Clinical Medicine, University of Bergen; the Department of Obstetrics and Gynecology, Haukeland University Hospital; the Medical Birth Registry of Norway, Norwegian Institute of Public Health; and the Locus for Registry Based Epidemiology, Department of Public Health and Primary Health Care, University of Bergen, Bergen, Norway.
See related article on page 16.
Funded by Norwegian Foundation for Health and Rehabilitation and the Norwegian Women's Public Health Association.
Corresponding author: Elham Baghestan, Department of Obstetrics and Gynecology Haukeland University Hospital, N-5021, Bergen, Norway; e-mail: email@example.com.
Financial Disclosure The authors did not report any potential conflicts of interest.
Obstetric anal sphincter injuries are common complications of vaginal delivery and can result in serious long-term problems; occurrence of anal incontinence after primary repair of obstetric anal sphincter injuries have been reported as high as 30% to 60%.1–3 There is a wide variation in reported risk of obstetric anal sphincter injuries, ranging from 1% to 11%.4–7 In recent years, an increase has been reported in some Scandinavian countries.7–9 However, it is unclear whether this increase is a result of improved diagnostic attention, more exact registration of cases, changes in obstetric routines, or increasing occurrence of other risk factors. In Norway, the increase has attracted much attention, and intervention programs have been introduced to reduce the risk of obstetric anal sphincter injuries.10
To prevent obstetric anal sphincter injuries, knowledge of risk factors, particularly modifiable, is necessary. Previous studies have focused on birth order of one, high birth weight, prolonged second stage of labor, and instrumental delivery,6,8,11 as well as African and Asian ethnicity.6,8 Episiotomy, epidural analgesia, and high maternal age also have been associated with obstetric anal sphincter injuries, but not consistently.6,12–14 Most previous studies have been based on small numbers or have investigated risk factors only in first births,8 whereas larger epidemiological studies on effect of birth order6,15 have not excluded births with previous obstetric anal sphincter injuries, which is necessary to avoid possible confounding attributable to a high recurrence of obstetric anal sphincter injuries.
The aim of the present study was to investigate risk factors for obstetric anal sphincter injuries in a large, population-based data set covering an extended period of time. In particular, we wanted to assess to what extent changes in risk factors over time could account for trends in obstetric anal sphincter injuries.
MATERIALS AND METHODS
Based on compulsory notification, the Medical Birth Registry of Norway has, since 1967, recorded data on maternal and neonatal health forwarded by the midwife and attending physician in a standardized form.16 The notification form was nearly unchanged until 1998, when a revised version was introduced.16 From 1967 through 2004, more than 2,000,000 births have been registered.
This registry-based cohort study included all women with a vaginal birth of a singleton and vertex-presenting fetus weighing 500 g or more. Women with their first birth before 1967 and births subsequent to obstetric anal sphincter injuries were excluded, leaving 1,673,442 births for study.
The dependent variable, obstetric anal sphincter injuries, was classified according to the International Classification of Diseases, 10th Revision (ICD-10) and included third-degree (ICD-10: O70.2), involving sphincter muscle, and fourth-degree (ICD-10: O70.3), involving rectal mucosa. From 1967 to 1998, obstetric anal sphincter injuries were reported to the Medical Birth Registry as plain text and coded as a dichotomous variable. The dependent variable has been validated with a satisfactory result.5 From 1999 onward, obstetric anal sphincter injuries have been reported as dichotomous variables by checking a box.
Independent variables comprised the following: year of delivery (1967–1976, 1977–1986, 1987–1996, 1997–2004); maternal age in years (younger than 20, 20–24, 25–29, 30–34, 35–39, 40 or older, unknown); and birth order, which was recorded as one, two, three, or four or greater, according to number of previous deliveries reported by the mother at birth. To assess the “pure” effect of a previous vaginal birth, we introduced the variable “vaginal birth order” based on the number of previous vaginal deliveries (1, 2, 3, 4, 5, 6, 7 or greater).
The deliveries were grouped according to history of vaginal birth or cesarean delivery as: 1) first vaginal delivery without previous cesarean delivery (first birth), 2) first vaginal delivery after one or more previous cesarean deliveries (previous cesarean only), 3) delivery after one or more cesarean and vaginal deliveries (previous vaginal and cesarean), and 4) delivery after one or more vaginal deliveries without previous cesarean delivery (previous vaginal only).
Other independent variables were: mother's country of birth (European, African, Asian, North American, Latin American, Oceanian, and unknown); diabetes (type 1, type 2, or gestational diabetes [yes or no]; data since 1999 were selected); smoking at the end of pregnancy (yes or no), recorded since 1999; mother's education in years (less than 8, 8–10, 11–12, 13–17, 18 or more, or unknown); size of maternity unit based on number of deliveries per year (less than 49, 50–499, 500–999, 1,000–1,999, 2,000–2,999, 3,000 or more); instrumental delivery (forceps, vacuum, both, noninstrumental); episiotomy (yes or no) recorded since 1999 (type is not specified; mostly mediolateral episiotomy is used in Norway); induction of labor by vaginal prostaglandin application (yes or no; data since 1999 were selected); epidural analgesia (yes or no); birth weight in grams (less than 2,500, 2,500–2,999, 3,000–3,499, 3,500–3,999, 4,000–4,499, 4,500–4,999, 5,000 or more); head circumference in centimeters (less than 33, 33–34, 35–36, 37–38, 39–40, 41 or more, unknown; data since 1980 were selected); and gestational age. Gestational age in weeks was estimated by subtracting the first day of last menstrual period from the date of birth. From 1999, gestational age based on ultrasound dating was available and was used when data on the last menstrual period were lacking (7.2%) (less than 37, 37–38, 39–40, 41–42, 43 or more, unknown).
Risks of a single woman were assessed in more than one pregnancy, and the data were thus organized in a two-level hierarchy with clusters of level-one data (the current pregnancy) nested within each level-two unit (the woman). To avoid underestimated standard errors caused by the nested structure of the data, associations of obstetric anal sphincter injuries with maternal and fetal characteristics and obstetric outcomes were assessed by multilevel logistic regression analysis. Adjusting was made for possible confounding variables, such as year of delivery, maternal age, birth order, and birth weight. The influence of these and other possible confounders on association between obstetric anal sphincter injuries and main exposures were tested by deviance statistics (P<.05). The statistical analyses were performed with SPSS (SPSS, Chicago, IL) and the MlWin program (Centre for Multilevel Modeling, University of Bristol, UK). The regional committee for medical research ethics approved the study protocol (REK Vest number 247.09).
The occurrence of obstetric anal sphincter injuries in vaginal vertex deliveries increased from 0.5% in 1967 to 4.1% in 2004. Obstetric anal sphincter injuries were more frequent in forceps and vacuum deliveries than in noninstrumental deliveries, and in all three categories the same trend was observed (Fig. 1). Obstetric anal sphincter injuries were more frequent in women with previous cesarean only and birth order of one than in vaginal birth order of two or more with the same trend (Fig. 2).
Also, trends in the independent variables were observed (Table 1). During the study period, the use of epidural analgesia increased in vaginal birth order one and vaginal birth order two or more (Table 1). The proportion of neonates weighing 4,000 g or more increased and was highest in vaginal birth order two or more. Median maternal age increased by 5 years in vaginal birth order of one and vaginal birth order of two or more. The use of forceps was almost unchanged but the use of vacuum increased in both groups. The proportion of women from Africa and Asia giving birth in Norway increased.
Table 2 shows frequencies as well as crude and adjusted odds ratios (ORs) of obstetric anal sphincter injuries for maternal and neonatal characteristics and obstetric interventions. Obstetric anal sphincter injuries were significantly associated with late period of delivery, maternal age older than 30 years, vaginal birth order of one, previous cesarean delivery, instrumental delivery, episiotomy, type 1 diabetes, gestational diabetes, induction of labor by prostaglandin, large maternity unit, birth weight 3,500 g or more, head circumference 35 cm or more, and African and Asian country of birth. No or marginal associations were observed with type 2 diabetes, gestational age, epidural analgesia, smoking, and mother's education.
The highest crude occurrence of obstetric anal sphincter injuries by maternal age was observed in mothers 25 to 34 years of age (Table 2). However, after adjusting for year of delivery, vaginal birth order, and birth weight, the association with maternal age was linear (Table 2). In logistic regression analysis, it appeared that vaginal birth order had the greatest effect. The effect of maternal age was independent of vaginal birth order (Fig. 3). After adjusting for several possible confounders, the increase of obstetric anal sphincter injuries persisted, although it reduced from 7.1-fold to 5.6-fold (Table 2).
There was a significant higher risk of obstetric anal sphincter injuries in women with previous cesarean delivery compared with women with no previous cesarean delivery (Table 2, Fig. 3). However, instrumental delivery was also more frequent in the former group. Still, in a stratified analysis of noninstrumental deliveries, we observed a significantly higher risk of obstetric anal sphincter injuries in the group with previous cesarean only compared with the group with first birth (OR 1.3; 95% confidence interval [CI] 1.2–1.4; Table 3), whereas in instrumental deliveries there was no significant difference in adjusted OR between the two groups (OR 1.1; 95% CI 0.97–1.2; Table 3). In noninstrumental deliveries, a higher risk of obstetric anal sphincter injuries was observed in the previous vaginal and cesarean groups compared with the previous vaginal-only group (OR 1.6; 95% CI 1.4–1.9; Table 3).
The effect of episiotomy was different in instrumental and noninstrumental deliveries. In first vaginal (birth order 1) deliveries, our data indicated no “protection” against obstetric anal sphincter injuries when episiotomy was used in noninstrumental deliveries (OR 1.0; 95% CI 0.97–1.1); however, in instrumental deliveries the use of episiotomy was “protective” against obstetric anal sphincter injuries (OR 0.8; 95% CI 0.76–0.9; Table 4). However, in second or higher vaginal (birth order 2+) deliveries, we found a higher risk of obstetric anal sphincter injuries when episiotomy was used in noninstrumental deliveries (OR 1.3; 95% CI 1.2–1.5) and no significant “protective” effect of episiotomy against obstetric anal sphincter injuries in instrumental deliveries (OR 0.8; 95% CI 0.6–1.1).
Risk of obstetric anal sphincter injuries was associated with size of maternity unit, even after additional adjusting for instrumental delivery (Table 2). The effect of maternal country of birth remained almost unchanged after additional adjusting for size of maternity unit.
Data from our national registry showed an increase in the number of obstetric anal sphincter injuries only partially accounted for temporal changes in observed risk factors. Important risk factors for obstetric anal sphincter injuries were vaginal birth order of one, high maternal age, instrumental delivery, previous cesarean delivery, and large fetus. Obstetric anal sphincter injuries were more frequent in women born in Asia and Africa and were associated with size of maternity unit, even after adjusting for the observed risk factors.
Strengths of our study are the population-based design and the prospective collection of data, reducing selection, and recall bias. Our database holds data on several possible confounders, and the outcome variable has been validated.5 Lack of data precluded assessing possible effects of events in the second stage of labor, like support of perineum, delivery position, the position and descent of the fetal head, and the skills of birth attendants.
As in previous reports,6,11,12,17,18 our study suggests that birth order of one is a dominant risk factor for obstetric anal sphincter injuries. Whereas previous vaginal delivery reduces the risk of obstetric anal sphincter injuries, our results suggest that previous cesarean delivery increases the risk, regardless of previous vaginal delivery. However, in instrumental delivery we found no significant excess risk associated with previous cesarean delivery, possibly because the effect of instrumental delivery or its indication on the risk of obstetric anal sphincter injuries is superior to the effect of the previous cesarean delivery. The risk of obstetric anal sphincter injuries in vaginal births subsequent to cesarean delivery has been studied before,6,11,17 but only one of these studies6 compared women with previous vaginal delivery with and without earlier cesarean delivery, as well. As in the present study, Handa and Richter6,17 reported higher risk estimates of obstetric anal sphincter injuries in women who had a previous cesarean delivery. However, another study11 reported no excess risk of obstetric anal sphincter injuries in women with a previous cesarean, possibly because they adjusted the effects for instrumental delivery. The excess risk after a previous cesarean delivery might be attributed to the indication of the cesarean delivery, which may persist in the next pregnancy.
The association between maternal age and risk of obstetric anal sphincter injuries has been suggested before, but not consistently.8,12,14 Regardless of birth order, the risk for obstetric anal sphincter injuries increased with maternal age until 30 years. In birth order three or more, the increase continued past the age of 30. A possible explanation could be less elasticity of the perineal tissue in older women.
In agreement with other studies,8,14,18,19 the risk of obstetric anal sphincter injuries was fourfold higher in forceps deliveries and twofold higher in vacuum deliveries compared with noninstrumental deliveries. The higher risk in forceps deliveries than in vacuum deliveries persisted after adjusting for possible confounders like year of delivery, maternal age, and vaginal birth order.
Midline episiotomy has been associated with obstetric anal sphincter injuries,20 but the association with mediolateral episiotomy, as generally used in Norway, is less consistent.12–14,18,21,22 The present study noted an overall increased risk of obstetric anal sphincter injuries associated with episiotomy. However, our results suggest that episiotomy in birth order one with noninstrumental delivery had no effect on obstetric anal sphincter injuries but had a “protective” effect in instrumental deliveries. Consistent with the present study, two studies have reported a “protective” effect of mediolateral episiotomy in instrumental vaginal delivery,21,23 but without distinction according to birth order. Additionally, because of lack of data on the recognized individual variation in the angle and size of mediolateral episiotomy,13,22,24 our results are not conclusive.
Women born in Africa and Asia had a higher risk of obstetric anal sphincter injuries than European women. Our finding that Asian women have a higher risk is supported by other studies.6,8,11,17 African women in our study population were mainly from Somalia, Eritrea, and Ethiopia, where 80% of women have been infibulated.25 This might account for the higher risk of obstetric anal sphincter injuries. One may speculate that the higher risk in African and Asian women is also caused by difficult communication with the birth attendants.
We found almost a twofold risk for obstetric anal sphincter injuries in maternity units with more than 3,000 births per year as compared with hospitals with 1,000 to 1,999 births per year. Consistent with our results, a recent Norwegian study26 found significant differences in the occurrence of obstetric anal sphincter injuries in noninstrumental deliveries between five hospitals and speculated that the differences could be caused by different perineal-protection handling techniques. Because larger maternity units in Norway are referral hospitals and treat more complicated instrumental deliveries and, consequently, may have higher risk of obstetric anal sphincter injuries, we adjusted the effect of size of maternity unit for instrumental delivery but the results remained.
Possible explanations for the increase of obstetric anal sphincter injuries in the study period include: 1) improved routines for the registration of obstetric anal sphincter injuries in the Medical Birth Registry of Norway, 2) improved diagnostic attention and routines, and 3) changes in obstetric and demographic risk factors. In a previous study, we assessed the validity of obstetric anal sphincter injuries in the Medical Birth Registry of Norway from 1990 to 1992 and from 2000 to 20025 and found a high validity of the Medical Birth Registry of Norway in both periods. The new notification form used from 1999 may have improved recording, but we observed no sudden change in the occurrence after the introduction of the new form (Figs. 1 and 2). Therefore, the new form likely has not significantly influenced the trend. Moreover, the “hands-off” practice promoted from the 1960s and 1970s27 may have contributed to the increased occurrence of obstetric anal sphincter injuries. One might also speculate that the concomitant shift in Norway from delivery unit-based to university college-based midwife training may have changed the management of the second stage.
During the past two decades, endoanal ultrasonography has revealed unrecognized or occult obstetric anal sphincter injuries. Studies have suggested that increased vigilance and appropriate examination,4,27,28 probably along with higher demands for documentation, may have improved the diagnostic.
The proportion of African and Asian women giving birth in Norway is increasing. The birth population is getting older and the proportions of heavy fetuses, use of epidural analgesia, and instrumental delivery have increased. However, after adjusting for such possible confounders, the trend in obstetric anal sphincter injuries persisted (Table 2), although our results indicate that some of the increase may be caused by these changes. In conclusion, changes in birth order, instrumental delivery, birth weight, previous cesarean delivery, and demographic risk factors can explain only some of the substantial increase of obstetric anal sphincter injuries during 1967 to 2004.
1. Norderval S, Nsubuga D, Bjelke C, Frasunek J, Myklebust I, Vonen B. Anal incontinence after obstetric sphincter tears: incidence in a Norwegian county. Acta Obstet Gynecol Scand 2004;83:989–4.
2. Zetterstrom J, Lopez A, Holmstrom B, Nilsson BY, Tisell A, Anzen B, et al. Obstetric sphincter tears and anal incontinence: an observational follow-up study. Acta Obstet Gynecol Scand 2003;82:921–8.
3. Sultan AH, Kamm MA, Hudson CN, Bartram CI. Third degree obstetric anal sphincter tears: risk factors and outcome of primary repair. BMJ 1994;308:887–91.
4. Andrews V, Sultan AH, Thakar R, Jones PW. Occult anal sphincter injuries-myth or reality? BJOG 2006;113:195–200.
5. Baghestan E, Bordahl PE, Rasmussen S, Sande AK, Lyslo I, Solvang I. A validation of the diagnosis of obstetric sphincter tears in two Norwegian databases, the Medical Birth Registry and the Patient Administration System. Acta Obstet Gynecol Scand 2007;86:205–9.
6. Handa VL, Danielsen BH, Gilbert WM. Obstetric anal sphincter lacerations. Obstet Gynecol 2001;98:225–30.
7. Laine K, Gissler M, Pirhonen J. Changing incidence of anal sphincter tears in four Nordic countries through the last decades. Eur J Obstet Gynecol Reprod Biol 2009;146:71–5.
8. Ekeus C, Nilsson E, Gottvall K. Increasing incidence of anal sphincter tears among primiparas in Sweden: a population-based register study. Acta Obstet Gynecol Scand 2008;87:564–73.
9. Medical Birth Registry of Norway. Annual reports 1967–2008. Medical Birth Registry of Norway, Bergen, and National Institute of Public Health, Oslo, Norway.
10. Laine K, Pirhonen T, Rolland R, Pirhonen J. Decreasing the incidence of anal sphincter tears during delivery. Obstet Gynecol 2008;111:1053–7.
11. Lowder JL, Burrows LJ, Krohn MA, Weber AM. Risk factors for primary and subsequent anal sphincter lacerations: a comparison of cohorts by parity and prior mode of delivery. Am J Obstet Gynecol 2007;196:344 e1–5.
12. Gerdin E, Sverrisdottir G, Badi A, Carlsson B, Graf W. The role of maternal age and episiotomy in the risk of anal sphincter tears during childbirth. Aust N Z J Obstet Gynaecol 2007;47:286–90.
13. Andrews V, Sultan AH, Thakar R, Jones PW. Risk factors for obstetric anal sphincter injury: a prospective study. Birth 2006;33:117–22.
14. Samarasekera DN, Bekhit MT, Preston JP, Speakman CT. Risk factors for anal sphincter disruption during child birth. Langenbecks Arch Surg 2009;394:535–8.
15. Spydslaug A, Trogstad LI, Skrondal A, Eskild A. Recurrent risk of anal sphincter laceration among women with vaginal deliveries. Obstet Gynecol 2005;105:307–13.
16. Irgens LM. The Medical Birth Registry of Norway. Epidemiological research and surveillance throughout 30 years. Acta Obstet Gynecol Scand 2000;79:435–9.
17. Richter HE, Brumfield CG, Cliver SP, Burgio KL, Neely CL, Varner RE. Risk factors associated with anal sphincter tear: a comparison of primiparous patients, vaginal births after cesarean deliveries, and patients with previous vaginal delivery. Am J Obstet Gynecol 2002;187:1194–8.
18. De Leeuw JW, Struijk PC, Vierhout ME, Wallenburg HC. Risk factors for third degree perineal ruptures during delivery. BJOG 2001;108:383–7.
19. Christianson LM, Bovbjerg VE, McDavitt EC, Hullfish KL. Risk factors for perineal injury during delivery. Am J Obstet Gynecol 2003;189:255–60.
20. Signorello LB, Harlow BL, Chekos AK, Repke JT. Midline episiotomy and anal incontinence: retrospective cohort study. BMJ 2000;320:86–90.
21. De Leeuw JW, de Wit C, Kuijken JP, Bruinse HW. Mediolateral episiotomy reduces the risk for anal sphincter injury during operative vaginal delivery. BJOG 2008;115:104–8.
22. Eogan M, Daly L, O'Connell PR, O'Herlihy C. Does the angle of episiotomy affect the incidence of anal sphincter injury? BJOG 2006;113:190–4.
23. Combs CA, Robertson PA, Laros RK Jr. Risk factors for third-degree and fourth-degree perineal lacerations in forceps and vacuum deliveries. Am J Obstet Gynecol 1990;163:100–4.
24. Tincello DG, Williams A, Fowler GE, Adams EJ, Richmond DH, Alfirevic Z. Differences in episiotomy technique between midwives and doctors. BJOG 2003;110:1041–4.
25. Elchalal U, Ben-Ami B, Gillis R, Brzezinski A. Ritualistic female genital mutilation: current status and future outlook. Obstet Gynecol Surv 1997;52:643–51.
26. Valbo A, Gjessing L, Herzog C, Goderstad JM, Laine K, Valset AM. Anal sphincter tears at spontaneous delivery: a comparison of five hospitals in Norway. Acta Obstet Gynecol Scand 2008;87:1176–80.
27. McCandlish R. Perineal trauma: prevention and treatment. J Midwifery Womens Health 2001;46:396–401.
© 2010 by The American College of Obstetricians and Gynecologists.
28. Groom KM, Paterson-Brown S. Can we improve on the diagnosis of third degree tears? Eur J Obstet Gynecol Reprod Biol 2002;101:19–21.