The incidence, risk factors and obstetric outcome in primigravid women sustaining anal sphincter tears

ORIGINAL ARTICLE

The incidence, risk factors and obstetricoutcome in primigravid women sustaininganal sphincter tearsNANDINI GUPTA, TALAKERE USHA KIRAN, VARSHA MULIK, JACKIE BETHEL AND KIRON BHAL

From the Department of Obstetrics and Gynecology, University of Wales College of Medicine, Cardiff, UK

Acta Obstet Gynecol Scand 2003; 82: 736–743. # Acta Obstet Gynecol Scand 82 2003

Background. The incidence of anal sphincter tears is highest among nulliparous women.The aim of this study was to ascertain if there were other factors that increased their risk.Methods. This was a retrospective study of all primigravid vaginal deliveries that hadsustained an anal sphincter tear (n¼ 122), compared with deliveries that did not have thiscomplication (n¼ 16 050). The study sample was drawn from a computerized maternityinformation database, comprising 52 916 deliveries in the South Glamorgan region during1990–99. SPSS version 10 was used for statistical analysis.Results. The incidence of anal sphincter tears in this study population was 0.8% (122/16172). Postdates (OR¼ 1.8, 95%CI¼ 1.3–2.6) and fetal macrosomia (OR¼ 3.8, 2.4–6)together with induction of labor (OR¼ 1.5, 1.01–2.2), use of spinal analgesia at delivery(OR¼ 3.1, 1.1–8.4), assisted vaginal delivery (OR¼ 1.9, 1.3–2.7; especially the use offorceps, OR¼ 2.2, 1.3–3.9) and doctor-conducted deliveries (OR¼ 2.2, 1.6–3.2) werefound to be associated with a significantly higher incidence of anal sphincter tears.Logistic regression revealed fetal macrosomia and doctor-conducted deliveries to beindependent risk factors that, when occurring together, were associated with a fourfoldincrease in the risk of occurrence of anal sphincter tears.Conclusions. This study suggests that careful assessment and counseling of women, par-ticularly >40 weeks gestation or those potentially having macrosomic fetuses, especially ifforceps are to be used for prolonged second stage in primigravid women, may help toidentify those at significant risk of anal sphincter tears.

Key words: anal sphincter tears; risk factors; primigravid women

Submitted 14 November, 2002Accepted 24 January, 2003

Anal incontinence is an extremely distressingcondition that affects 20–50% of women follow-ing anal sphincter rupture during vaginal delivery(1). The incidence of anal sphincter rupture variesbetween 0.5 and 2.5% in centers where mediolat-eral episiotomy is practiced, compared with 7%in units performing midline episiotomies (2–4).

The incidence is highest among nulliparouswomen (2,5) and those undergoing assistedvaginal delivery. Among many other factors con-tributing to rectal injury, episiotomy is the mostcontroversial and has been the subject of manystudies. A number of other factors may contrib-ute to rectal injury, including induction, aug-mentation, duration of labor, use of anesthesia,birthweight and the operator conducting thedelivery (6).

In order to prevent these tears we must have aclear understanding of the potential causativefactors. As primigravid women are at a higher

Abbreviations:PPH: postpartum hemorrhage; ARM: artificial rupturemembrane; BMI: body mass index; SPSS: Statistical Packagefor Social Sciences.

# Acta Obstet Gynecol Scand 82 (2003)

Acta Obstet Gynecol Scand 2003: 82: 736--743 Copyright # Acta Obstet Gynecol Scand 2003

Printed in Denmark. All rights reservedActa Obstetricia et

Gynecologica ScandinavicaISSN 0001-6349

risk of anal sphincter tears this study aims toidentify possible causative factors in primigravidwomen and provide recommendations for itsprevention. It also aims to assess the effect ofthese tears on postnatal health.

Materials and methods

This was a retrospective analysis of obstetric vari-ables in 122 primigravid women who had sus-tained an anal sphincter tear, compared with theremaining 16 050 primigravid vaginal deliverieswithout such a complication in the same period.

The study sample was drawn from CardiffBirths Survey, a population-based database com-prising 52 916 women who had delivered in theSouth Glamorgan region during 1990–99. Thedatabase is robust and is regularly audited withappropriate training for the staff who input datainto the database. Information from this data-base has been used in other studies published inpeer review journals. Furthermore, the data thatthis study was based on were entered into aMaternity Information System that was rigid inits design and as such had restrictions regardingthe data that could be entered into each field.There was cross-field checking and this includedmessages to ensure that coders double checkedany information that was felt to be ‘unusual’, forexample out of normal range. The deliveries werechecked on a regular basis with labor wards,cross-checking of numbers and outcome. When-ever the Births Survey uses the data, furthercrosschecking is carried out and, where possible,anything that looks unlikely is investigated. If theoriginal data cannot be retrieved, clinical adviceis sought and, if necessary, a delivery would beexcluded from a piece of analysis.

This study comprised primigravid womenundergoing vaginal delivery. The exclusion cri-teria other than multiparity included stillbirths,fetal congenital malformations, multiple preg-nancies and cesarean sections. The total numberof women included in the study after strict appli-cation of the inclusion and exclusion criteria was16 172.

The data were divided into two groups: a groupof deliveries without anal sphincter tears (deliv-eries without any tears, those with episiotomyand first- or second-degree perineal tears) and agroup of deliveries with anal sphincter tears. Inthe study population, maternal characteristicssuch as age and body mass index (BMI), gesta-tional age (with postdates �40 weeks of gesta-tion), labor characteristics such as induction,augmentation, duration of labor, conductionalanalgesia, interventions during labor and fetal

characteristics such as birthweight were analyzedas potential risk factors for anal sphincter tears.Secondary outcome measures were assessed,including maternal morbidity and perinatal mor-bidity and mortality.

Definitions used in this study

An anal sphincter tear was defined as any tearinvolving the anal sphincter muscles with or with-out involvement of the anal mucosa. This isbecause the database did not classify the tearsseparately; for example third-degree tears (3a,3b and 3c: less than 50% of external anal sphinc-ter torn, more than 50% of external anal sphinc-ter torn, and internal sphincter torn, respectively,according to Sultan’s (7) classification) andfourth-degree tears (torn anal epithelium aswell). Delayed second stage was defined as beinggreater than 2 h from full dilatation of the cervix.Postpartum hemorrhage (PPH) included primaryand secondary PPH. Postdates was defined asgestational age greater than 40 weeks. Macro-somia was defined as birthweight greater than4000 g. Induction methods included prostaglan-dins, artificial rupture of membranes (ARM) withor without oxytocin. Augmentation includedARM or oxytocin or both.

Statistical analysis

SPSS version 10 (Statistical Package for SocialSciences; SPSS Inc, Chicago, IL, USA) was usedfor statistical analysis. Unpaired samples t-test,�2-test and Fisher’s exact test were used whereappropriate. A p-value of less than 0.05 was con-sidered significant. An odds ratio (OR) with a95% confidence interval (CI) was used to evalu-ate the strength of association between outcomemeasures. The possible effect of confoundingvariables was controlled using logistic regressionanalysis.

Results

The overall incidence of anal sphincter tears inthe study group was 0.8% (122/16172). Thematernal characteristics, pregnancy and deliverydetails are described in Table I. There appears tobe no difference in the age and BMI data betweenthe two cohorts studied. Mean gestational ageand, hence, incidence of postdates and inductionrate were significantly greater in women who hadan anal sphincter tear compared with those with-out this complication. The incidence of macro-somia and the use of spinal analgesia at deliverywere other variables found to be significantly

Anal sphincter tears in primigravid women 737


higher in women who subsequently ended upwith an anal sphincter tear. Although the use ofepidural analgesia for labor or delivery or lengthof second stage was not different between the twogroups in Table I, the use of epidural analgesiadid prolong the mean duration of the secondstage in this study population (epidural 123� 68min vs. no epidural 66� 47 min; unpaired t-test;

p< 0.001). The intrapartum details between thetwo groups were otherwise comparable.

The mean birthweight and incidence of macro-somia were both significantly higher in the groupof women with anal sphincter tears (Table I).This, however, did not affect the incidence ofshoulder dystocia in either group of women(Table II). Deliveries conducted by medical staff

Table I. Maternal characteristics, pregnancy and delivery details of the study population

Anal sphincter tearn¼ 122

No anal sphincter tearn¼ 16050 Odds ratio (OR) or significance

Maternal characteristicsAge (years) 26.3� 5.6 25.4� 5.5 NSBMI 23.9� 4.4 23.8� 4.2 NS

Pregnancy detailsGestational age (weeks) 40.2� 1.3 39.6� 1.8 **Incidence of postdates 51 (41.8) 4530 (28.2) **OR1.8 (1.3–2.6)

Delivery detailsInduction of labor 34 (27.9) 3271 (20.4) *OR1.5 (1.01–2.2)Prostaglandins 21 (17.2) 2008 (12.5)Oxytocin 7 (5.7) 880 (5.5)ARM 2 (1.6) 126 (0.8)All of the above 4 (3.3) 257 (1.6)

Augmentation 54 (44.3) 6460 (40.2) NSAnalgesiaEpidural only for labor and delivery 44 (36.1) 5334 (33.2) NSEpidural in labor with spinal for delivery 1 (0.8) 37 (0.2) NSSpinal only for delivery 4 (3.3) 175 (1.1) *OR3.1 (1.1–8.4)

Duration of first stage (min) 461� 233 443� 237 NSDuration of second stage (min) 92� 65 87� 62 NSIncidence of delayed second stage (>2h) 31 (25.4%) 3979 (24.8%) NSBirthweight (g) 3488� 492 3281� 497 **Incidence of macrosomia 25 (20.5) 1015 (6.3) **OR3.8 (2.4–6.0)Lead personnel at deliveryDoctor 51 (41.8) 3902 (24.3) **OR2.2 (1.6–3.2)Midwife 71 (58.2) 12148 (75.7) NS

Figures are mean values� SD or numbers/incidence with percentage in parentheses.NS, not significant (p> 0.05); *p< 0.05; **p< 0.001.Unpaired t-test was used to compare continuous variables. �2- or Fisher’s exact test was used for proportions.

Table II. Postnatal maternal and neonatal outcome measures

Anal sphincter tearn¼ 122

No anal sphincter tearn¼ 16050

Odds ratio (OR) orsignificance

MaternalDuration of postnatal stay (days) 3�1.8 2.7� 1.7 *Incidence of:Postpartum hemorrhage 4 (3.3) 290 (1.8) NSPostnatal blood transfusion 12 (9.8) 427 (2.7) **OR 4 (2.2–7.3)Perineal infection 1 (0.8) 37 (0.2) NS

NeonatalDuration of hospital stay (days) 3� 1.9 3� 4.6 NSIncidence of:Birth trauma 9 (7.4) 883 (5.5) NSAsphyxia 0 14 (0.1) NS5min Apgar score <6 3 (2.5) 146 (0.9) NSShoulder dystocia 2 (1.6) 106 (0.7) NS

Neonatal Unit admissions 3 (2.5) 675 (4.2) NSPerinatal mortality (all neonatal deaths) 1 (0.8) 28 (0.2) NS

Figures are mean values� SD or numbers/incidence with percentage in parentheses.NS, not significant (p> 0.05); *p< 0.05; **p< 0.001.Unpaired t-test was used to compare continuous variables. �2- or Fisher’s exact test was used for proportions.

738 K. Bhal et al.


were associated with a twofold increase in theincidence of anal sphincter tears compared withthose conducted by midwives (Table I).

Figure 1 depicts the difference in mode ofdelivery in this study population. Assisted vaginaldelivery (OR¼ 1.9, 1.3–2.7), in particular the useof forceps (OR¼ 2.2, 1.3–3.9), was strongly asso-ciated with anal sphincter tears. However, only1.6% (36/2311) of forceps delivery and 0.7% (18/2546) of ventouse deliveries resulted in analsphincter tears. A comparison of mean birth-weight between instrumental and spontaneousvertex deliveries showed that the cohort of babiesbeing delivered by forceps or ventouse was sig-nificantly heavier (weight 3388� 472 g vs.3243� 496 g; p< 0.05; unpaired t-test).

Figure 2 shows the trends in the episiotomyand anal sphincter tears rates over the 10-yearstudy period. The usual practice in this region isto perform a mediolateral episiotomy. However,if a midline episiotomy were to have been

carried out by an operator then this wouldnot necessarily be coded differently in this data-base. During the study period there was aninitial reduction followed by stabilization of theoverall episiotomy rate (mean 40%) without anappreciable increase in the anal sphincter tearrate.

Table II shows that the mean length of stay formothers was significantly longer in those deliv-eries complicated by anal sphincter tears.Although the incidence of postpartum hemor-rhage was not significantly increased in womenwith anal sphincter tears, these deliveries wereassociated with an almost fourfold increase inpostnatal blood transfusion. This discrepancycould be due to the inaccuracy of estimatingblood loss at vaginal delivery. The perineal infec-tion rates were not significantly different betweenthe two groups. The neonatal outcome measureswere not significantly different between thegroups.

0

10

20

30

40

50

60

70

80

Women with anal sphinctertears (n = 122)

Women without analtears (n = 16050)

sphincter

Mode of delivery

%

SVD

AVD

Breech

Ventouse

Forceps

n = 67

n = 54 n =4803

n = 1n = 18

n =11140

n = 107

n = 2528n = 2275

*

*

n = 36

Fig. 1. Differences in mode of deliverywithin the study population. SVD,spontaneous vertex delivery; AVD,assisted vaginal delivery; *p< 0.05(SVD or AVD: anal sphincter tears vs.no anal sphincter tears).

0

10

20

30

40

50

60

70

90 91 92 93 94 95 96 97 98 99

Year of delivery (1990–1999)

Inci

den

ce (

%)

anal sphincter tear rate (mean = 0.8%)

episiotomy rate (mean = 40%)

Fig. 2. The trends in episiotomy andanal sphincter tear rates over the10-year study period within the studypopulation (n¼ 16 172).



All variables that were found to be significantlydifferent between the two study cohorts weresubjected to logistic regression (Table III). Thisrevealed that macrosomia and lead personnel atdelivery were the only independent variablesaffecting the incidence of anal sphincter tears.Therefore, if both these factors were present inthe same delivery the odds of having an analsphincter tear was calculated to be 4.4.

Discussion

The reason for choosing to study risk factors inprimigravid women is because they are at highestrisk of anal sphincter tears and this reduces otherpotential confounding variables from beingincluded in any retrospective analysis. To thebest of our knowledge this is the largest studyassessing risk factors specifically in primigravidwomen.

The overall reported incidence regardless ofparity is 0.5 to 2.5% in centers practicing medio-lateral episiotomy (2,3). It is usually consideredthat primigravid women are at a higher risk ofsustaining these injuries, which may be due tothe relative inelasticity of the perineum (2). Con-sidering these factors, the incidence in this studypopulation of 0.8% was relatively low. This couldof course be due to the possibility of missingoccult anal sphincter tears as two prospectivestudies have shown that up to one-third ofwomen undergoing their first vaginal deliverysustain anal sphincter damage that is not recog-nized at delivery (8,9).

Some risk factors, namely increasing gesta-tional age at delivery (especially �40 weeks),induction of labor, use of forceps, birthweight(especially �4000 g) and doctor-conducted deliv-ery, found to be associated with anal sphinctertears in this study are confirmed in previousreports (2,6,12–14). However, after logisticregression only the latter two risk factors werefound to be independently associated with analsphincter tears.

Donnelly et al. compared results of a bowelfunction questionnaire and anal vector mano-metry before and 6 weeks after delivery in aprospective cohort study of 184 primiparouswomen (9). These authors found that instru-mental delivery and a second stage of laborprolonged by epidural analgesia are the obstet-ric factors posing the greatest risk of injury tothe anal sphincter mechanism (9). This studyfound instrumental delivery and not delayedsecond stage or epidural analgesia to be a riskfactor for anal sphincter tear in primigravidwomen.

Increased maternal age (>30 years) has beenquoted to be a risk factor in some studies on thebasis of the hypothesis that aging leads todecreased elasticity of the perineum and thusincreases the propensity to tear (10). However,no association with age was demonstrated inthis study.

Other authors have found that increasinggestational age had no effect on the risk of analsphincter tears in instrumental deliveries, suggest-ing that it was the instrument that was the import-ant risk factor (11). However, the converse wasfound in this study. Interestingly, in this study,increased gestational age was associated withincreased incidence of induction of labor in thecohort with anal sphincter tears. It is difficult toascertain whether this was a cause-and-effectsituation, especially when considering gestationalage and induction as possible risk factors. Therewas no difference in the outcome regardless of themethod used for induction, or for augmentation.The potential mechanism implicating inducedlabor as a risk factor leading to anal sphinctertear is thought to be due to increased pressure onthe perineum (12,13). In a prospective study bySamuelson et al. (13), univariate analysis revealedthat the use of oxytocin was a risk factor, but itwas no longer significant after stepwise logisticregression. In this study there was no associationwith oxytocin usage.

Two studies have suggested that birthweighthas a stronger influence in increasing the inci-dence of anal sphincter tears in nulliparous thanin parous women (12,13). Others have docu-mented a stepwise (OR¼ 1.47, 1.43–1.51) increasein the risk of third-degree tears with each 500-gincrease in birthweight (14). This study concurswith the importance of birthweight as a risk fac-tor as increased mean birthweight, especiallymacrosomia, was found to be significantly higheramong women with anal sphincter tears. Macro-somia remained an independent variable evenafter logistic regression analysis. In contrast to astudy by Moeller Bek and Laurberg, no association

Table III. Outcome of logistic regression analysis on variables that potentiallyaffect the incidence of anal sphincter tears in this study

Variable Odds ratio and 95% CI

Gestational age (weeks) 1.3 (1.0–1.6)Postdates 0.8 (0.5–1.5)Induction of labor 1.3 (0.8–1.9)Spinal analgesia alone at delivery 2.0 (0.8–5.0)Assisted vaginal delivery 0.7 (0.3–1.6)Incidence of macrosomia 2.8 (1.8–4.6)*Lead personnel at delivery (doctor) 2.6 (1–6)*

*Statistically significant.

740 K. Bhal et al.


was found between shoulder dystocia and analsphincter tears (5).

Studies have suggested that epidural analgesiaprolongs the second stage of labor and increasesthe risk of anal sphincter tears during labor (9)especially in primigravid women (12). Coombset al. suggested that epidural analgesia decreasesthe risk of anal sphincter tears during instrumen-tal delivery by relaxing the perineal musculature(11). In this study epidural analgesia significantlyprolonged the second stage of labor, but therewas no association with use of epidural analgesiaand incidence of anal sphincter injuries.However, the difference in the use of spinalanesthesia may well be due to the possibility ofneeding regional anesthesia to repair these analsphincter injuries as the database did not differ-entiate between the use of regional anesthesia fordelivery or complications of the third stage oflabor.

Duration of second stage of labor has beenreported as unrelated to the risk of third-degreetear in the past (6,15). In the study by MoellerBek and Laurberg (5), the mean duration of secondstage was a risk factor, with an unadjusted oddsratio of 4.06 (2.5–6.6) that was adjusted to1.6 (0.9–2.3) in the multiple regression logisticanalysis and was no longer significant. Anotherstudy suggested that it was only the duration ofbearing down that remained significant afterlogistic regression whereas the duration of theentire second phase was not an independentfactor of importance (13). The duration of secondstage was not a significant factor in this studypopulation. However, the duration of the bearingdown period was not coded separately in theCardiff Births Survey and was therefore notassessed in this study. de Leeuw et al. found theduration of second stage to be significantly asso-ciated (OR¼ 1.12, 1.10–1.14) with third-degreeperineal tears per 15-min increase in the lengthof the second stage (14). This again was not repli-cated in this population.

Poen et al. suggested that a mediolateral episi-otomy may have a protective role especially inthe primigravid population (12), while othershave found an almost threefold increase in therisk of anal sphincter tears when this interventionwas used (5). Sultan et al. found that 42% ofwomen who sustained a third-degree tear withoutan instrumental delivery did so despite postero-lateral episiotomy (2). Reynolds found that redu-cing the episiotomy rate among the primiparouswomen (from 57.6% to 46.2%; p< 0.001) wasassociated with a significant decrease in the rate ofthird- and fourth-degree perineal tears (16). How-ever, data from this population over the 10-year

period showed an initial reduction followed bya plateau in the episiotomy rate without anappreciable increase in the third-degree tear rate,suggesting that episiotomies may not necessarilyhave an effect on the occurrence or preventionof anal sphincter tears. This finding appearsto confer with that of Henriksen et al. (17)that change in the use of episiotomy, namely areduction in its rate, was not associated with aconcomitant rise in tears of the anal sphincter.Henriksen et al. (17) also suggested that the over-all episiotomy rates should ideally be kept below30%, in comparison to which the episiotomy ratein this study is relatively high (mean 40%),although Williams et al. (18) found that there isan appreciable variation in episiotomy ratesthroughout regions and hospitals in the UK,ranging from 26 to 67%. This could be becausethis is a highly selective population, namely onlyprimigravid women, where episiotomy rates evenafter modifying clinical practice were up to 46%(in primiparous women) in the study publishedby Reynolds (16).

In the study by Sultan et al., half the womenwho sustained third-degree tears were deliveredby forceps, although this complication occurredonly in 4% of the forceps deliveries (2). Theassociation with ventouse delivery was less cer-tain (19). Where an attempt at ventouse deliveryfails and conversion to forceps occurs, there issignificant increase in the risk of anal sphinctertear (5,14,20). In this study, a primigravid womanundergoing an assisted vaginal delivery wasalmost twice as likely (OR¼ 1.9, 1.3–2.7) tosustain an anal sphincter tear, in particular withthe use of forceps rather than ventouse. A random-ized prospective study of 600 women found ahigher incidence of maternal injuries with forceps(21), supporting the view that the ventouseshould be the instrument of choice (22). A recentstudy implicated vacuum extraction (both outletand mid release) as a independent risk factor andsuggested that the option of cesarean sectioninstead of vacuum extraction should be con-sidered when mid release is needed in the presenceof macrosomia (23).

Surprisingly, instrumental delivery was notfound to be an independent risk factor, probablybecause only 1.6% forceps delivery and 0.7% ofventouse deliveries resulted in anal sphinctertears. A further comparison of mean birthweightbetween instrumental and spontaneous vaginaldeliveries showed that the cohort of babiesbeing delivered by forceps or ventouse was sig-nificantly heavier implying that instrumentaldeliveries per se may not contribute directly toanal sphincter tears.



This is one of the more recent studies assessingthe outcome in relation to the status of the leadperson at delivery. Green and Soohoo (6) foundthat presence of a physician at delivery was animportant factor in deliveries complicated by analsphincter tears with an adjusted OR of2.4, 1.6–3.6 (doctor vs. midwife). However, asinstrumental deliveries increase the risk of analsphincter tears, it would not be surprising thatdelivery by physicians was found to be associatedwith increased risk of anal sphincter tear in thisstudy.

A recent study by Janni et al. found no correla-tion between delayed second stage and analsphincter tears in multivariate analysis and alsosuggested that the increased maternal morbidityin patients with prolonged labor may be partiallyattributed to the higher rate of operative inter-ventions (24). We concur and suggest that aneffort should be made to minimize instrumentaldeliveries for prolonged second stage where pos-sible. Perhaps a randomized controlled trial of alonger passive second stage prior to the com-mencement of the active stage in primigravidwomen could help to determine whether a reduc-tion in instrumental deliveries and similarly inanal sphincter tears could be achieved in prim-iparous women without a worsening of neonataloutcome.

In conclusion, this study identified fetal macro-somia and doctor-conducted deliveries as inde-pendent risk factors that, when occurringtogether, were associated with a fourfold increasein the likelihood of an occurrence of anal sphinc-ter tears. Therefore, when counseling regardingthe risks of induced labor, especially in poten-tially macrosomic pregnancies for postdates inprimigravid women, consideration should begiven to include the increased risk of anal sphinc-ter tears under these circumstances.

Acknowledgments

We thank the Cardiff Births Survey for their assistance inthe completion of this project.

References

1. Sultan AH, Monga AK, Kumar D, Stanton SL. Pri-mary repair of obstetric anal sphincter rupture usingoverlap technique. Br J Obstet Gynaecol 1999; 106:318–23.

2. Sultan AH, Kamm MA, Hudson CN, Bartram CL.Third degree obstetric anal sphincter tears: risk factorsand outcome of primary repair. Br Med J 1994; 308:887–91.

3. Tetzschner T, Sorensen M, Lose G, Christianen J. Analand urinary incontinence in women with obstetric anal

sphincter rupture. Br J Obstet Gynaecol 1996; 103:1034–40.

4. Klein MC, Gauthier RJ, Robbins JM, Kaczorowski J,Jorgensen SH, Franco ED et al. Relation of episiotomyto perineal trauma and morbidity, sexual dysfunctionand pelvic floor relaxation. Am J Obstet Gynecol 1994;171: 591–8.

5. Moeller Bek K, Laurberg S. Intervention during labour:risk factors associated with complete tear of theanal sphincter. Acta Obstet Gynecol Scand 1992; 71:520–4.

6. Green JR, Soohoo SL. Factors associated with rectalinjury in spontaneous deliveries. Obstet Gynecol 1989;73: 732–8.

7. Sultan AH. Obstetrical perineal injury and anal incon-tinence. Clin Risk 1999; 5: 193–6.

8. Sultan AH, Kamm MA, Hudson CN, Thomas JM,Bartram CI. Anal-sphincter disruption during vaginaldelivery. N Engl J Med 1993; 329: 1905–11.

9. Donnelly V, Fynes M, Campbell D, Johnson H,O’Connell PR, O’Herlihy C. Obstetric events leadingto anal sphincter damage. Obstet Gynecol 1998; 92:955–61.

10. Fynes M, O’Herlihy C. The influence of mode ofdelivery on anal sphincter injury and fecal continence.[Review.] Obstetrician Gynaecologist 2000; 3: 120–5.

11. Combs CA, Robertson PA, Laros RK Jr. Risk factorsfor third degree and fourth degree lacerations in forcepsand vacuum deliveries. Am J Obstet Gynecol 1990; 163:100–4.

12. Poen AC, Felt-Bersma RJ, Dekker GA, Deville W,Cuesta MA, Meuwissen SG. Third degree obstetricperineal tears: risk factors and the preventive role ofmediolateral episiotomy. Br J Obstet Gynaecol 1997;104: 563–6.

13. Samuelsson E, Ladfors L, Wennerholm UB, Garebeg B,Nyberg K, Hagberg H. Anal sphincter tears: a prospect-ive study of obstetric risk factors. Br J Obstet Gynaecol2000; 107: 926–31.

14. de Leeuw JW, Struijk PC, Vierhout ME, Wallenburg HC.Risk factors for third degree perineal rupturesduring delivery. Br J Obstet Gynaecol 2001; 108:383–7.

15. Wilcox LS, Strobino DM, Baruffi G, Dellinger WS Jr.Episiotomy and its role in the incidence of perineallacerations in a maternity centre and a tertiary hospitalobstetric service. Am J Obstet Gynecol 1989; 160:1047–52.

16. Reynolds JL. Reducing the frequency of episiotomiesthrough a continuous quality improvement program.Can Med Assoc J 1995; 153: 275–82.

17. Henriksen TB, Bek KM, Hedegaard M, Secher NJ.Methods and consequences of changes in use of episiot-omy. Br Med J 1994; 309: 1255–8.

18. Williams FLV, Florey C, Mires GJ, Ogston SA. Episi-otomies and perineal tears in low-risk UK primigravidae.J Public Health Med 1998; 20: 422–7.

19. Sultan AH, Kamm MA, Bartram CI, Hudson CN. Analsphincter trauma during instrumental delivery. A com-parison between forceps and vacuum extraction. IntJ Gynaecol Obstet 1993; 43: 263–70.

20. Fitzpatrick M, Fynes M, Cassidy M, Behan M,O’Connell PR, O’Herlihy C. Prospective study of theinfluence of parity and operative technique on theoutcome of primary anal sphincter repair followingobstetrical injury. Eur J Obstet Gynecol Reprod Biol2000; 89: 159–63.

21. Johanson RB, Rice C, Doyle M, Arthur J, Anyanwu L,Ibrahim J et al. A randomized prospective studycomparing the new vacuum extractor policy with

742 K. Bhal et al.


forceps delivery. Br J Obstet Gynaecol 1993; 100:524–30.

22. Chalmer JA, Chalmer I. The obstetric vacuum extractorshould be the instrument of choice for operativevaginal delivery. Br J Obstet Gynaecol 1989; 96:505–6.

23. Jander C, Lyrenas S. Third and fourth degree perinealtears. Predictor factors in a referral hospital. ActaObstet Gynecol Scand 2001; 80: 229–34.

24. Janni W, Schiessl B, Peschers U, Huber S, Strobi B,Hantschmann P et al. The prognostic impact ofa prolonged second stage of labor on maternal and

fetal outcome. Acta Obstet Gynecol Scand 2002; 81:214–21.

Address for correspondence:Kiron BhalDepartment of Obstetrics and GynecologyLlandough HospitalPenlan RoadCardiff CF64 2XXWalesUKe-mail: [email protected]



Documents

The incidence, risk factors and obstetric outcome in primigravid women sustaining anal sphincter tears