Hals, Elisabeth RN; Øian, Pål MD, PhD; Pirhonen, Tiina RN; Gissler, Mika DrPhil, MSOCSCI; Hjelle, Sissel MD; Nilsen, Elisabeth Berge MD; Severinsen, Anne Mette RN; Solsletten, Cathrine RN; Hartgill, Tom MD; Pirhonen, Jouko MD, PhD
Long-lasting anal incontinence symptoms are present in 30–50% of women with a clinically recognized third- or fourth-degree perineal laceration despite adequate primary repair at delivery.1–4 The complications after obstetric anal sphincter injury may be devastating, and therefore reducing obstetric anal sphincter injury rates and the associated sequelae by raising standards of clinical practice is an important step in improving the quality of obstetric care.
Seeking to decrease the number of obstetric anal sphincter injury, several studies have been conducted with varying results. Improving delivery technique,5 changing obstetric practice by recommending the use of vacuum extraction instead of forceps, mediolateral instead of medial episiotomy, instructing the mother to push with less effort while the fetal head crowns,6 and the use of manual support of the perineum7–9 have been shown to be protective against obstetric anal sphincter injury. Some smaller randomized studies have shown no significant difference whether manual support was used or not.10–13
In Norway, we have experienced a gradual increase in the incidence of obstetric anal sphincter injury from under 1% in the late 1960s to 4.3% in 2004 (Medical Birth Registry of Norway). After a review of the Norwegian Board of Health in 2004, a national strategy to reduce the number of obstetric anal sphincter injury was developed. As part of this program, an interventional study was commissioned for the national strategy. Based on our positive short-term (1.5-year) results from an interventional study in Fredrikstad, Norway,9 we decided to extend the project to four new hospitals, in which the design focus was on 1) good communication between the accoucheur and the delivering woman; 2) adequate perineal support; 3) a delivery position that allows visualization of the perineum during the last minutes of delivery; and 4) episiotomy only on indication.
The aim of this study was to estimate whether such an intensive interventional program would lead to a sustained reduction in anal sphincter tears when expanded into a multicenter setting.
MATERIALS AND METHODS
The present project includes four delivery units, one from each of the four health regions in Norway. Two of the hospitals are university hospitals (Tromsø and Stavanger) and two county hospitals (Lillehammer and Ålesund). Stavanger is a larger delivery unit than the other three. Lillehammer had the highest national frequency of obstetric anal sphincter injury 3 years before intervention. In the 3 years before the intervention, the total obstetric anal sphincter injury frequency was clearly higher than the overall obstetric anal sphincter injury frequency (4.3%) in Norway in three of the four hospitals, ie, 4.84% (Lillehammer), 4.75% (Stavanger), and 4.98% (Ålesund), but lower in Tromsø at 3.68%.
In total, 40,152 vaginal deliveries between 2003 and 2009 were enrolled in this interventional cohort study. Basic data from the four delivery units are presented in Table 1.
The hospitals have some variation in their obstetric practice. In Stavanger and Ålesund, there is a tradition for the use of an episiotomy resembling a median episiotomy. Lillehammer and Stavanger have a tradition for using forceps mostly, whereas Tromsø and Ålesund use a vacuum extractor as the preferred instrument. Before the intervention, the use and technique in the perineal support varied greatly in every participating clinic.
The clinical intervention consisted of actively training labor ward personnel starting with a 2- or 3-day course of compulsory tutorials at each participating delivery unit. The national project leader conducted the tutorials. The program included the basics of anal sphincter tears and a presentation of the ongoing project. The entire labor ward staff took part. The physicians were also instructed in the use of similar manual protective techniques for use with vacuum extractors or forceps. Two to four sessions with similar programs at each hospital were arranged to give all midwives and doctors the possibility to participate in the tutorials.
The aim was to focus on the four aspects as stated at the beginning of this article and to reintroduce the midwives and doctors to the traditional method of assisting delivery of the neontae during the final part (crowning) of the second stage of delivery, a procedure commonly used in Norway before the 1980s. The delivery assistant presses the neonate's head with the left hand to control the speed of crowning through the vaginal introitus, and the right hand supports the perineum and tries to take a grip on the neonate's chin. When a good grip has been achieved, the woman is asked to stop pushing and to breathe rapidly while the midwife slowly helps the neonate's head through the vaginal introitus. When most of the head is out, the perineal ring is pushed under the neonate's chin.
Delivery room personnel were instructed to encourage the women to move freely during delivery and push in the position they felt most comfortable. Furthermore, they should help the women to choose a position during the last minutes of pushing such that the classic maneuver could be performed while allowing the perineum to be observed.
We also focused on the use of episiotomy. Routine episiotomy was not recommended, except when clinically indicated on an individual basis. If an episiotomy was performed, the cut should be far away from the anal opening (so-called lateral or mediolateral episiotomy) to avoid obstetric anal sphincter injury.
One of the main goals for the training period was to establish a local core team of experts for each unit, who should continue training and continuous reassessment of the staff after the midwife instructor had completed the active training period. These midwives and doctors were exposed to more deliveries than the rest of the staff with the explicit aim to achieve superior competence.
Tromsø was first to start the practical supervision in January 2006 followed by Lillehammer, Ålesund, and Stavanger. The practical training in Stavanger was completed in May 2007. The practical training period was conducted by an experienced midwife from Finland. The training period varied from 7 to 13 weeks depending on the number of deliveries at the participating hospitals. All members of the staff took part in the training. They were instructed and supervised in three stages: first a practical session using a pelvic model; then the midwife or doctor was supervised in the technique in a clinical setting. Initially, the instructing midwife had her hands over the accoucheur's to teach the correct technique. Finally, the midwives or doctors were allowed to deliver under supervision.
After the training program was instituted, neither the national project leader nor the midwife responsible for the initial practical training and tutorials became involved in the training or everyday practice of the units. The local core team of expert staff carried on the intervention locally.
Data from all deliveries in the four participating hospitals were collected from the obstetric databases from at least 3 years before the intervention until the end of June 2009. The length of data collection after the intervention was 3 years 6 months in Tromsø, 3 years in Lillehammer, 2 years 9 months in Ålesund, and 2 years 3 months in Stavanger. To minimize bias involved with recording administrative data, we manually checked and validated the data at the time of database enrollment. In addition, for all cases of obstetric anal sphincter injury, the original operation notes and history were assessed by the authors aimed at correctly classifying the grade of obstetric anal sphincter injury.
The grade of obstetric anal sphincter injury was determined as follows: 3a, less than 50% of the external anal sphincter depth torn; 3b, more than 50% of the external anal sphincter depth torn; 3c, internal anal sphincter torn with or without torn external sphincter; and 4, injury of the anal sphincter complex and anorectal epithelium.2,9 The obstetric anal sphincter injury diagnosis was done by the midwife or doctor who was responsible for the delivery. The diagnosis was always confirmed by a specialist in obstetrics and gynecology.
For statistical analysis, the proportion of parturients with anal sphincter rupture during delivery (International Classification of Diseases, 10th Revision, code O70) was calculated per 100 vaginal births. Their share was compared by type of delivery (instrumental and noninstrumental) and by timing (before and after intervention) using the test for relative proportions. Similar statistical methods were used to analyze for possible confounding factors. P<.05 was considered statistically significant. Furthermore, for change in the total obstetric anal sphincter injury frequency as well as for noninstrumental and instrumental deliveries, the odds ratios and 95% confidence intervals were calculated. The study was approved by the Regional Committee North for Medical and Health Research Ethics.
The total proportion of parturients with obstetric anal sphincter injuries decreased from 4.16–5.25% before intervention to 1.73% during the last year of intervention (Table 2). The overall obstetric anal sphincter injury rate differed significantly from preinterventional rates already 1 year after the start (P<.001) (Fig. 1).
Lillehammer and Tromsø maintained a constant decrease in the obstetric anal sphincter injury rates with a final year rate of 1.29% in Lillehammer and 1.21% in Tromsø, respectively (Fig. 1). The obstetric anal sphincter injury rate in Stavanger remained stable after the first year and then stayed at approximately 2% (Fig. 1). In Ålesund, the initial large decrease in obstetric anal sphincter injury showed a slight but nonsignificant increase during the second and the third years of intervention (P=.52 and P=.30, respectively).
A dramatic decrease of obstetric anal sphincter injury was observed for both noninstrumental (Table 3) and instrumental (Table 4) deliveries after the intervention started, and the overall drop was statistically significant for both groups already after the first year of intervention (P<.001) (Fig. 2A–B).
For noninstrumental deliveries, the decrease in obstetric anal sphincter injury was significant after the first year of intervention in all but one hospital, Lillehammer (P=.791). During the second, third, and fourth years, the drop was significant in all hospitals (P<.001). At the end of the registration period, the lowest obstetric anal sphincter injury frequency for noninstrumental deliveries was 0.62% (Lillehammer) and highest 2.13% (Ålesund) (Table 3; Fig. 2A). For instrumental deliveries, the lowest obstetric anal sphincter injury frequency during the last year of follow-up was 5.3% (Tromsø) and the highest 8.1% (Stavanger) (Table 4; Fig. 2B).
In Table 5, the data displaying changes in obstetric anal sphincter injury grades over the whole study period are presented. There is a statistically significant decrease in all obstetric anal sphincter injury grades (P<.001) compared with rates before the intervention (except 3c after the first year P=.042). Looking at the entire period before and after the intervention, the greatest decrease was in grade 4 tears (−63.5%), whereas the proportion of 3c tears decreased the least (−47.5%).
The number of episiotomies performed during the whole study period is presented in Table 6. The episiotomy frequency increased significantly (P<.001) both in Stavanger and in Ålesund. This change remained significant even at the end of the study period. In Lillehammer, no significant change in episiotomy rates took place. In Tromsø, the number of episiotomies increased slightly after the start of the intervention (first year P=.012, nonsignificant; second year P<.001), but then the trend reversed with the rate falling to a lower level than before the intervention (P=.75) (Table 6).
The number of instrumental deliveries (both vacuum and forceps together) before and after the intervention remained stable in Lillehammer and Tromsø and increased slightly in Ålesund and significantly in Stavanger (Table 1). The number of forceps was unchanged in Lillehammer and in Stavanger, respectively (Table 1). There were a few forceps deliveries in Ålesund and Tromsø before the intervention; the forceps rate decreased further after the start of the intervention (Table 1). No rotational forceps were performed. The number of deliveries by vacuum extraction increased significantly in Stavanger and Ålesund.
In the study period, no statistically significant changes occurred with regard to gestational age at delivery, mean birth weight (Table 1), shoulder dystocia, fetal presentation, body mass index before pregnancy (Tromsø and Lillehammer), maternal age at delivery, or maternal weight at delivery (Stavanger and Ålesund). Intervention had no deleterious effects on the neonates. The 5-minute Apgar score increased slightly, and there were no changes in the blood gases. The number of nulliparous women was higher after the start of intervention in Ålesund, epidural analgesia rates were higher in Ålesund, and the induction of labor rate increased in all clinics except Tromsø. The cesarean delivery rate decreased in Tromsø and increased in Stavanger (Table 1).
The results from the present study show a powerful effect of the intervention program causing a dramatic decrease in the total number of obstetric anal sphincter injury. Three of the hospitals had an obstetric anal sphincter injury frequency close to 5% before the intervention. By the end of the follow-up postintervention, all the hospitals had a total obstetric anal sphincter injury frequency between 1% and 2%. Furthermore, the most serious damage (grade 4) decreased more than grade 3 obstetric anal sphincter injury.
Frankman et al14 reported recently the incidence of obstetric anal sphincter injury for operative deliveries in the United States to lie between 14% and 16% and in spontaneous vaginal deliveries 4–5%. After the present intervention, the participating clinics have an obstetric anal sphincter injury frequency for operative vaginal deliveries close to the frequency found for spontaneous vaginal deliveries in the United States.
Our study from 19987 compared the obstetric anal sphincter injury rates in two university hospitals in Finland and Sweden and showed that the traditional methods of helping the newborn through the last stage of delivery, which is still a standard technique in Finland, could in part be an explanation for Finland's low obstetric anal sphincter injury frequency. Our hypothesis was first put to the test in an interventional study in Fredrikstad, Norway, 2002–2007,9 where reintroducing the method led to a decrease in overall obstetric anal sphincter injury frequency from 4.0% to 1.2%.
We preferred an interventional approach instead of a randomized controlled trial in our study. Several changes to clinical practice in our study are complex procedures that are poorly amenable to the methodology of large multicenter randomized trials. Our earlier results from the same type of intervention9 have shown a dramatic decrease in obstetric anal sphincter injury, and we did not find it ethical to provide women with an alternative we already knew was a poor one. Furthermore, similar techniques have been associated with an obstetric anal sphincter injury frequency under 1% in Finland through many decades. Therefore, a randomization, based on our experience, would have raised an ethical dilemma for us, so it was natural to choose an interventional model with a rigorous prospective design and data collection to minimize biases. However, use of this type of design makes the total control of a time course effect continuing through the intervention difficult as a result of a lack of a reference group.
The national focus on reduction of anal sphincter tears started in November 2004 after the report from the Norwegian Board of Health.15 Local and national information meetings and courses were arranged, and the traditional method was demonstrated on a pelvic mannequin. Debates for and against the technique have been fruitful with many delivery units trying to reduce the number of tears with their own initiatives. Overall in Norway, the frequency of obstetric anal sphincter injury has fallen from 4.3% in 2004 to 3.0% in 2007 (Medical Birth Registry of Norway).16 The four hospitals taking part in our intervention and the hospital in Fredrikstad have significantly lower and sustained obstetric anal sphincter injury rates compared with the national figures, although the initial obstetric anal sphincter injury rates were above the national average before the intervention.
In Ålesund, the proportion of obstetric anal sphincter injuries for instrumental deliveries remained constant at a low level, whereas the tears from noninstrumental deliveries rose, although not significantly. In Stavanger, the situation was reversed; the tears associated with vacuum extraction or forceps increased slightly. This may be the result of absence (maternity leave, for example) of key personnel, ie, midwives in Ålesund and obstetricians in Stavanger. It has been shown that inexperienced birth attendants may increase the perineal damage rate.17
The local expert teams' operative work differed from hospital to hospital and it is very likely that the explanation for the mentioned progress could be found here. The intervention started in Tromsø and Lillehammer. At that time, there was widespread scepticism toward the reintroduction of an old technique previously used in Norway but no longer practiced. The expert teams and heads of departments faced much criticism and were directly opposed by some birth attendants. However, step by step, the expert teams created a sound and well-functioning system. The very positive results were in themselves conductive to a change in attitudes and atmosphere at the units, because the “new” method soon became popular among the majority of midwives and physicians. Furthermore, birth attendants who experienced a sphincter tear completed a proforma as part of the clinical follow-up in the study to analyze reasons for the tears. In all hospitals, one of the main challenges has been to give new coworkers as well as locum staff the skills needed to avoid sphincter tears.
There were small changes in the traditional risk factors for obstetric anal sphincter injury,2,18,19 although none can explain the significant reduction in obstetric anal sphincter injury. Three of four hospitals experienced a clear increase in the number of delivery inductions, one hospital had a statistically significant increase in nulliparity, and one increased the number of operative deliveries. It is possible that the decrease in obstetric anal sphincter injury in our study would have been more pronounced without these changes.
The use of episiotomy has been considered a risk factor for obstetric anal sphincter injury, but results from published studies are contradictory.20 The majority of recent studies have demonstrated an increased incidence of third- and fourth-degree tears associated with the use of midline episiotomy. Based on data from the United States, particularly with regard to operative vaginal deliveries, midline episiotomy is strongly associated with an increased risk for obstetric anal sphincter injury,21,22 and nonmidline episiotomy should be used to protect the anal sphincters.23 It has been shown that an increase in distance between the anus and episiotomy and an increase in the angle of episiotomy away from the anal opening will cause a significant reduction in the risk of obstetric anal sphincter injury.24,25
As a result of our project, the number of anal sphincter ruptures in participating delivery clinics is now close to the levels seen during the late 1960s. Therefore, we advise other hospitals to introduce an interventional program as described here aimed at reducing the morbidity of this serious complication of vaginal delivery.
1.Haadem K, Dahlstrom J, Lingman G. Anal sphincter function after delivery: a prospective study in women with sphincter rupture and controls. Eur J Obstet Gynecol Reprod Biol 1990;35:7–13.
2.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.
3.Norderval S, Øian P, Revhaug A, Vonen B. Anal incontinence after obstetric sphincter tears: outcome of anatomic primary repairs. Dis Colon Rectum 2005;48:1055–61.
4.Borello-France D, Burgio K, Richter H, Zyczynski H, Fitzgerald MP, Whitehead W, et al. Fecal and urinary incontinence in primiparous women. Obstet Gynecol 2006;108:863–72.
5.Parnell C, Langhoff-Roos J, Moller H. Conduct of labor and rupture of the sphincter ani. Acta Obstet Gynecol Scand 2001;80:256–61.
6.Hirsch E, Haney EI, Gordon TE, Silver RK. Reducing high-order perineal laceration during operative vaginal delivery. Am J Obstet Gynecol 2008;198:668.e1–5.
7.Pirhonen JP, Grenman SE, Haadem K, Gudmundsson S, Lindqvist P, Siihola S, et al. Frequency of anal sphincter rupture at delivery in Sweden and Finland—result of difference in manual help to the baby's head. Acta Obstet Gynecol Scand 1988;77:974–7.
8.Samuelsson E, Ladfors L, Wennerholm UB, Gåreberg B, Nyberg K, Hagberg H. Anal sphincter tears: prospective study of obstetric risk factors. BJOG 2000;107:926–31.
9.Laine K, Pirhonen T, Rolland R, Pirhonen J. Decreasing the incidence of anal sphincter tears during pregnancy. Obstet Gynecol 2008;111:1053–7.
10.McCandlish R, Bowler U, van Asten H. A randomised controlled trial of care of the perineum during second stage of normal labour. Br J Obstet Gynaecol 1988;105:1262–72.
11.Mayerhofer K, Bodner-Adler B, Bodner K, Rabl M, Kaider A, Wagenbichler P, et al. Traditional care of the perineum during birth. A prospective, randomized, multicenter study of 1,076 women. J Reprod Med 2002;47:477–82.
12.Albers LL, Sedler KD, Bedrick EJ, Teaf D, Peralta P. Midwifery care measures in the second stage of labor and reduction of genital tract trauma at birth: a randomized trial. J Midwifery Womens Health 2005;50:365–72.
13.Jönsson ER, Elfaghi I, Rydhström H, Herbst A. Modified Ritgen's maneuver for anal sphincter injury at delivery. A randomized controlled trial. Obstet Gynecol 2008;112:212–7.
14.Frankman EA, Wang L, Bunker CH, Lowder JL. Episiotomy in the United States: has anything changed? Am J Obstet Gynecol 2009;200:573.e1–7.
16.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.
17.Jandér C, Lyrenäs S. Third and fourth degree perineal tears. Predictor factors in referral hospital. Acta Obstet Gynecol Scand 2001;80:229–34.
18.Zetterström J, Lopez A, Anzen B, Norman M, Holmström B, Mellgren A. Anal sphincter tears at vaginal delivery: risk factors and clinical outcome of primary repair. Obstet Gynecol 1999;94:21–8.
19.Dudding TC, Vaizey CJ, Kamm MA. Obstetric anal sphincter injury: incidence, risk factors, and management. Ann Surg 2008;247:224–37.
20.Woolley RJ. Benefits and risks of episiotomy: a review of the English-language literature since 1980. Part I. Obstet Gynecol Surv 1995;50:806–20.
21.Helwig JT, Thorp JM Jr, Bowes WA Jr. Does midline episiotomy increase the risk of third- and fourth-degree lacerations in operative vaginal deliveries? Obstet Gynecol 1993;82:276–9.
22.Kudish B, Blackwell S, Mcneeley SG, Bujold E, Kruger M, Hendrix SL, et al. Operative vaginal delivery and midline episiotomy: a bad combination for the perineum. Am J Obstet Gynecol 2006;195:749–54.
23.de Leeuw J, de Wit C, Bruinse H, Kuijken J. Mediolateral episiotomy reduces the risk for anal sphincter injury during operative vaginal delivery. BJOG 2008;115:104–8.
24.Andrews V, Sultan AH, Thakar R, Jones PW. Risk factors for obstetric anal sphincter injury: a prospective study. Birth 2006;33:117–22.
25.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.
© 2010 by The American College of Obstetricians and Gynecologists.