Risk Factors for Anal Sphincter Tear in Multiparas.

Abstract


OBJECTIVE:

To assess maternal, newborn, and obstetric risk factors associated with anal sphincter tear in multiparous women.

METHODS:

This case–control study identified 18,779 multiparous vaginal deliveries from 1992 to 2004 from an obstetric automated record database at the University of Alabama at Birmingham. Two hundred eighty-four patients were selected, 145 cases and 139 controls. Variables from the index pregnancy and prior pregnancies were analyzed, and multivariable logistic regression models were constructed to determine significant predictor variables for anal sphincter tear in multiparous women.

RESULTS:

One hundred forty-five multiparous women with no history of cesarean delivery sustained a sphincter tear. Multivariable logistic regression showed a significant association with episiotomy, shoulder dystocia, forceps delivery, and being married. A second exploratory model that included variables from previous pregnancies, showed that in addition to episiotomy, shoulder dystocia , forceps delivery, previous sphincter tear, and second stage of labor greater than 1 hour were associated with tear.

CONCLUSION:

The strongest clinical risk factors for anal sphincter tear in multiparous women are episiotomy, shoulder dystocia, previous sphincter tear, prolonged second stage of labor, and forceps delivery.

Introduction

Pelvic floor morbidity, including urinary incontinence (UI), fecal incontinence (FI), and prolapse, has been associated with vaginal delivery. Prevalence of stool incontinence has been reported to be 3% after vaginal birth, while the rate of flatal incontinence was 37%. Stool incontinence is even more common after an anal sphincter tear, affecting 8–40% of women. Multiparous women have a higher risk of urinary and fecal incontinence symptoms than primiparous women, as many have already sustained injury to the pelvic floor with a previous delivery.

Risk factors commonly cited for obstetric sphincter tear are primiparity, episiotomy, macrosomia, operative delivery, and shoulder dystocia. The prevalence of anal sphincter tears in general is approximately 6% but has been reported to be as high as 10%. However, fewer studies have reported the occurrence of sphincter tears in multiparous women, in whom rates range from 1% to 4%.

Risk factors associated with sphincter tear in primiparous deliveries are reasonably well characterized, but only seven studies have specifically addressed risk factors for sphincter tears in the multiparous patient. These studies have been large cohort studies using databases with limited variables. Most have revealed an increased risk of anal sphincter tears with a history of previous third- and fourth-degree tear but have not detailed other risk factors.

DISCUSSION

We have shown that a number of maternal, labor, and newborn variables of both the index and previous pregnancies may be associated with an increased risk of anal sphincter tear in the multiparous woman. Many reports have documented risk factors for sphincter tear in the nulliparous patient. Interestingly, the factors associated with multiparous sphincter tears are similar to those of the nulliparous patient, including forceps delivery, prolonged second stage of labor, and episiotomy. This may reflect common labor practices for both the nulliparous and parous women.

The current literature describing risk factors for sphincter tear in multiparous women includes seven other studies. The most common risk factors identified were a history of prior tear, forceps delivery, episiotomy, and birth weight more than 4,000 g, which is consistent with risk factors identified in our study. In addition, Spyslaug et al also identified prolonged labor greater than 24 hours or second stage of labor greater than 60 minutes as a risk factor for sphincter tear, and Elfaghi et al found increasing maternal age as risk factor for tear.

Episiotomy was the risk factor most strongly correlated with sphincter tear in our study. Episiotomy has not been found to be protective to the pelvic floor as had been previously thought, and its use should not be routine. As with episiotomy, forceps delivery was also very strongly associated with sphincter tear. Operative delivery, while often indicated, is a risk factor for sphincter tear, and practitioners may consider counseling patients before labor about operative vaginal delivery and its possible sequelae. Shoulder dystocia, while not predictable, often requires increased force on the pelvic floor or an episiotomy or proctoepisiotomy performed to extricate the fetus at the expense of potential traumatic perineal injury.

Because a woman has a “proven pelvis” does not mean that she is at decreased risk for a traumatic delivery. Obstetricians may consider counseling multiparous patients that a history of sphincter tear is a risk factor for a subsequent tear as found in our study and multiple others with subsequent increased risk for symptoms of fecal incontinence. Whether an elective cesarean delivery should be offered to these patients is unclear and controversial, as primary cesarean itself has not been shown to be totally protective against symptoms of fecal and urinary incontinence.

Increasing length of second stage of labor as a risk factor for sphincter tear may be reflective of our propensity to hasten delivery by operative delivery or with the use of episiotomy. Two retrospective studies have shown that increased rates of operative deliveries, postpartum hemorrhage, and chorioamnionitis are associated with second stage of labor longer than 2 hours. Therefore, practitioners should consider not only the aforementioned maternal morbidity but also an increased risk of sphincter tear when counseling multiparous patients regarding continued pushing versus cesarean delivery during a prolonged second stage of labor.
Being married was associated with sphincter tear in multiparous women, a finding which has not been report in other studies. It is possible that being married is a marker for higher socioeconomic class, maternal age, or other factors not represented in our models. Of note, with the models adjusted for increased birth weight and maternal age, being married was still a significant risk factor. Future studies may need to more fully evaluate characteristics associated with being married.

This study differs from other multiparous risk factor studies in that we evaluated multiple maternal, fetal, and labor factors of the index and previous pregnancies. This is a heterogenous population including 51% African American, 40% white, and 4% Hispanics as compared with the most recent large studies of homogenous populations from Sweden and Norway. Furthermore, all variables were confirmed by a thorough chart review to alleviate potential inherent bias.

Limitations of this study include those associated with the use of an automated database where there may be miscoding at the time of data entry thereby missing or falsely identifying sphincter tears and other covariates; however, the charts of all index pregnancies and available prior pregnancies were reviewed by hand to minimize any discrepancies. Fifty percent of the past pregnancy charts were missing, which may have had an impact on the results.

In summary, multiparous patients with a history of vaginal delivery with an anal sphincter tear may be at risk for a recurrent tear. Furthermore, having an episiotomy, shoulder dystocia, second stage of labor greater than or equal to 1 hour, and forceps in a subsequent delivery may be associated with an increased risk of sustaining a tear. Recommendations for future pregnancy management would include minimizing the use of episiotomy and operative delivery with forceps. However, well-controlled prospective clinical studies are needed to determine if that would result in decreased risk for sphincter tear.