Obstetric anal sphincter injuries occur in 0.25–7.31% of vaginal deliveries1,2 and can result in severe morbidity, including anal incontinence, perineal pain, dyspareunia, and emotional distress.3 Understanding modifiable risk factors for severe perineal trauma and identifying strategies to minimize them are important for long-term patient outcomes, especially as life expectancy increases and women live longer in the aftermath of obstetric anal sphincter injury.
Attempts to minimize pelvic floor trauma during delivery have resulted in obstetric practice shifting away from forceps-assisted vaginal delivery toward vacuum-assisted vaginal delivery as a result of purported higher rates of obstetric anal sphincter injury and pelvic floor injury among patients with forceps-assisted vaginal delivery (8–23%)4–6 compared with vacuum-assisted vaginal delivery (6–9%).7–10 A recent study found that graduates in 2003 recorded an average of 23.8 forceps-assisted vaginal deliveries during their residencies compared with 8.4 forceps-assisted vaginal deliveries recorded by graduates in 2013, a 64.7% decline (P<.001).11 Little is known about the effect of this shift toward vacuum as the operative delivery instrument of choice on the rate of obstetric anal sphincter injury.
Additionally, modern obstetric practice supports a longer duration of the second stage of labor in an effort to reduce rising cesarean delivery rates. Recommendations by the American College of Obstetricians and Gynecologists to prevent primary cesarean delivery were updated in 2014 to include no specific limit to the second stage, stating that longer labor duration is appropriate as long as progress is documented.12 However, studies have indicated increased obstetric anal sphincter injury risk with a prolonged second stage.13,14
In this study, we described delivery outcomes reflective of current obstetric practice in a diverse Northern California managed care organization and characterized the obstetric anal sphincter injury rate and risk factors. We aimed to identify subpopulations at uniquely elevated risk of obstetric anal sphincter injury.
MATERIALS AND METHODS
This is a retrospective cohort study of 22,822 vaginal deliveries within the Kaiser Permanente Northern California health care system from January 2013 to December 2014. Kaiser Permanente Northern California is a managed health care delivery system that serves 4.14 million people within Northern California, both publically (MediCal) and privately insured, and includes 13 labor and delivery units across the region. All units are staffed by in-house obstetricians with support of midwives who collaborate with attending obstetricians. Of these 13 units, three units are primary sites of residency training programs in obstetrics and gynecology. The selected time interval facilitated a large sample size while being reflective of contemporary practices. After receiving approval by the Kaiser Permanente institutional review board, we identified all live, term (37 weeks of gestation or greater), singleton, cephalic vaginal deliveries that occurred over the 2-year study period. The database allowed follow-up of 22,822 of 31,522 (72.4%) patients. Women were included if they were continuously insured through Kaiser Permanente from delivery until 6 months postpartum to allow tracking of postpartum complications. Demographic and clinical characteristics of the excluded cohort did not differ from the women included in the study except that those who were excluded were more likely to be Hispanic (28% vs 24%, P<.005). Using electronic medical records, we extracted demographic and health information (maternal age, race or ethnicity defined by patient's self-identification, maternal prepregnancy body mass index [BMI, calculated as weight (kg)/[height (m)]2], and parity) as well as delivery outcome characteristics (duration of the second stage of labor, epidural status, vaginal birth after cesarean delivery [VBAC] status, episiotomy status, delivery type, and neonatal birth weight). Because fewer than 0.4% of women (n=81) had a forceps-assisted vaginal delivery, we excluded forceps-assisted vaginal delivery from our final analyses, resulting in a total cohort of 22,741 deliveries. A sensitivity analysis done before this exclusion did not alter the results.
Reliance on the electronic medical record minimized missing data points. The key data points for our study were identified as duration of the second stage of labor and perineal laceration severity; when either of these data points was missing, women were excluded from the analysis. Figure 1 displays the flow diagram outlining the final patient cohort included in our study.
The primary outcome was the incidence rate of obstetric anal sphincter injury. Obstetric anal sphincter injury was defined as either 1) a third- or fourth-degree perineal laceration that was documented in the delivery flowsheet or 2) International Classification of Diseases, 9th Revision codes describing obstetric anal sphincter injury (569.43, 664.2, 664.3, 664.6) found in the Kaiser Permanente Northern California Medical Diagnoses database. Third- and fourth-degree lacerations were defined as in the July 2016 American College of Obstetricians and Gynecologists Practice Bulletin, “Prevention and Management of Obstetric Lacerations at Vaginal Delivery.”15 In cases of discrepancy between the delivery flowsheet record and International Classification of Diseases, 9th Revision codes (n=17), manual chart review was performed to ascertain the severity of the perineal laceration.
We used bivariate analyses—t tests for continuous variables, χ2 tests for categorical variables, and nonparametric tests when assessing differences in medians—to identify maternal demographic and health characteristics (age, race–ethnicity, BMI, parity) and delivery characteristics (duration of second stage of labor, epidural, VBAC, and episiotomy status; delivery type; birth weight) that were associated with obstetric anal sphincter injury. We then used multiple logistic regression analyses to control for known or suspected confounders, including parity and birth weight, to identify independent risk factors for obstetric anal sphincter injury. The crude and adjusted association between the duration of second stage of labor, categorized into four 60-minute intervals (0–59, 60–119, 120–179, 180 minutes or greater), and obstetric anal sphincter injury risk was assessed.
To better characterize the relationships between longer duration of the second stage and obstetric anal sphincter injury, we grouped duration of the second stage into 15-minute intervals (0–210 minutes) to examine the magnitude of effect conferred by these incremental increases. As a result of the small number in the greater than 210-minute category, we grouped them together. All univariate, bivariate, and multivariate analyses were conducted using SAS 9.3.
Our study cohort (N=22,741) had a mean age of 30.1 years (SD 5.5) and reflected the racial and ethnic diversity of Northern California with 24.0% identifying as Hispanic or Latina, 6.7% as African American, 24.1% as Asian, and 41.2% as white or Caucasian. The mean BMI was 25.8 (SD 5.6) and nearly two thirds of the women were multiparous (62.2%). The spontaneous vaginal delivery rate in our cohort was 93.8% (n=21,341), and 6.2% (n=1,400) delivered by vacuum-assisted vaginal delivery. The overall incidence rate of obstetric anal sphincter injury was 4.9% (n=1,111). In bivariate analyses (Table 1) obstetric anal sphincter injury incidence was higher among women of Asian obstetric anal sphincter injury, lower BMI, and nulliparas and lower among Hispanic women. A higher incidence of obstetric anal sphincter injury was also found among women who had an epidural (5.2%), VBAC (8.7%), episiotomy (21.3%), and vacuum-assisted vaginal delivery (24.0%). The mean birth weight of neonates of mothers with obstetric anal sphincter injury was 3,488.5 g (SD 464.3) vs 3,416.8 g (SD 439.3) in mothers without obstetric anal sphincter injury (P<.001) (Table 1).
The final multiple logistic regression model (Table 2) had good discrimination (c statistic 0.81) and calibration (Hosmer-Lemeshow P=.157). In adjusted analyses, most associations found in bivariate analyses persisted. However, having an epidural placed women at lower odds of obstetric anal sphincter injury. Compared with the group of women whose duration of the second stage of labor was less than 60 minutes, elevated odds of obstetric anal sphincter injury were associated with longer times (60–119 minutes: adjusted odds ratio [OR] 1.93, 95% CI 1.60–2.34; 120–179 minutes: adjusted OR 2.69, 95% CI 2.19–3.30; 180 minutes or greater: adjusted OR 3.20, 95% CI 2.62–3.89). Although Hispanics had a lower odds of obstetric anal sphincter injury compared with other racial and ethnic groups in our cohort (adjusted OR 0.72, 95% CI 0.58–0.90), Asians continued to be at higher risk (adjusted OR 2.31, 95% CI 1.99–2.69). Nulliparous women were significantly more likely to incur obstetric anal sphincter injury than multiparas (adjusted OR 2.32, 95% CI 2.00–2.71). Vacuum-assisted vaginal delivery was the risk factor that conferred the highest odds of obstetric anal sphincter injury (adjusted OR 4.23, 95% CI 3.59–4.98). Compared with the 3.6% of spontaneous deliveries with obstetric anal sphincter injury, 24.0% of women who delivered by vacuum experienced obstetric anal sphincter injury (P<.001). Women who had a mediolateral (adjusted OR 1.72, 95% CI 1.14–2.60) or midline (adjusted OR 2.93, 95% CI 2.36–3.65) episiotomy had a greater odds of obstetric anal sphincter injury compared with those with no episiotomy. Women delivering by VBAC had nearly three times the odds of obstetric anal sphincter injury compared with the general population (adjusted OR 2.87, 95% CI 2.14–3.85).
The duration of the second stage of labor in our population was reflective of increasingly permissive modern obstetric practices with a range of 0–680 minutes (Fig. 2). The overall median duration of the second stage of labor in our cohort was 48 minutes (interquartile range 17–114 minutes). For nulliparous women the median duration was 95 minutes (interquartile range 47–163); for multiparous women, the median was 27 minutes (interquartile range 12–73). Women with obstetric anal sphincter injury had a second stage of labor that was, on average, twice as long as those without obstetric anal sphincter injury (144.2 vs 74.6 minutes, P≤.001). This difference was more pronounced when we compared median duration of the second stage of labor, which was 129 minutes in women with obstetric anal sphincter injury as compared with 45 minutes in those without (P<.001) (Table 1).
A scatterplot of the duration of the second stage against the risk of obstetric anal sphincter injury was used to plot the trend line for the effect of duration of second stage of labor on the incidence of obstetric anal sphincter injury. The trend line depicts a nearly linear relationship between obstetric anal sphincter injury risk and the duration of the second stage of labor in 15-minute intervals (Fig. 3). There was no clear inflection point beyond which an incremental increase in obstetric anal sphincter injury risk for all women was observed.
Nearly one fourth of women (24.0%) delivered by vacuum incurred obstetric anal sphincter injury. This is a substantial increase over the historically quoted 6–9% obstetric anal sphincter injury rates with vacuum-assisted vaginal delivery and at least equal to 8–23% rates quoted with forceps-assisted vaginal deliveries.4–6 Whereas previous studies suggest that forceps-assisted vaginal delivery confers the highest obstetric anal sphincter injury risk,16–19 our findings indicate that operative delivery in general, rather than the instrument itself, is associated with the elevated obstetric anal sphincter injury rate. Historically, obstetricians may have reserved forceps for challenging deliveries, whereas contemporary obstetric practice favors using vacuum for virtually all operative deliveries. The high rate of obstetric anal sphincter injury with vacuum-assisted vaginal delivery in our results may reflect removal of this instrument selection bias. Corroboration of our findings by geographically diverse obstetric units would strengthen our conclusions.
The wide distribution of duration of the second stage reflects increasingly permissive labor management patterns. Duration of the second stage was independently associated with obstetric anal sphincter injury, even after controlling for vacuum-assisted vaginal delivery and VBAC. Labor management strategies aimed at shortening the second stage such as upright pushing positions and manual rotation of the malpositioned fetus20,21 may also decrease the need for operative delivery, thus reducing obstetric anal sphincter injury incidence.
Women with VBAC had nearly three times the odds of obstetric anal sphincter injury (8.7% incidence) compared with the general population. We could not ascertain the indication for cesarean delivery before VBAC, which limits our ability to draw conclusions from these data. However, VBAC was associated with higher obstetric anal sphincter injury rates in all subgroups of our population. Currently, discussion of VBAC includes counseling about risks of uterine rupture, emergent surgery, and their effects on the neonate; patients are not informed about elevated obstetric anal sphincter injury risk. Discussion of this risk must be coupled with a discussion of the benefits of a successful VBAC and the surgical and abnormal placentation risks associated with multiple cesarean deliveries.
The numerous risk factors for obstetric anal sphincter injury should not be viewed in isolation. Their complex interactions place certain subpopulations at significantly higher risk than other groups. It is important to recognize racial and ethnic differences in the incidence of adverse events and outcomes to begin investigating the underlying causes of such differences.
Our study has several strengths. Our diverse population, derived from a large managed care organization accounting for approximately 4.5% of 500,000 annual deliveries in California,22 makes our results generalizable to many practice environments. Our data were entered by clinicians at the time of delivery reducing the likelihood of coding, recall, and data entry errors. Our results reflect current obstetric practice patterns in the United States with 13 labor and delivery units employing hundreds of health care providers from geographically diverse training programs, different training philosophies (allopathic, osteopathic, midwifery), a wide range of years in practice, and functioning with and without residents.
Our study is limited by its retrospective design, which required health care providers and coders to accurately summarize clinical scenarios. An obstetrician's choice of delivery instrument or decision to allow a prolonged second stage in select patients could introduce selection bias. We were unable to ascertain indications for operative delivery except by manual chart review of clinician notes, which was not feasible with our sample size. Duration of the second stage was defined as the time period between complete cervical dilation and delivery; we were unable to delineate how much of this was active pushing time compared with laboring down. Our ability to stratify obstetric anal sphincter injury risk by delivering provider type was limited by the medical record, which does not automate residents' or midwives' involvement in the labor course. This limitation is mitigated by the organization of our labor and delivery units, which follow a hospitalist system with the attending obstetrician present for each delivery and comanaging patients with midwives on units with a midwifery service. Our study results differ from previously published data, implicating vacuum in the development of obstetric anal sphincter injury to a much higher extent than previously believed. Prospective studies examining the effect of delivery mode pelvic floor anatomy and long-term functional outcomes are needed to draw definitive conclusions about best practices.
As women have fewer children, live longer, and expect a higher quality of life in older age, long-term consequences of vaginal and cesarean delivery should be considered. Our study suggests the cost of achieving vaginal delivery varies based on the constellation of patients' demographic, health, and delivery characteristics. Shared decision-making efforts to establish patient labor management goals based on informed consent should consider not only immediate surgical risks of and maternal morbidity associated with cesarean delivery, but also long-term sequelae of pelvic floor injury.
1. Stock L, Basham E, Gossett DR, Lewicky-Gaupp C. Factors associated with wound complications in women with obstetric anal sphincter injuries (OASIS). Am J Obstet Gynecol 2013;208:327.e1–6.
2. Dandolu V, Chatwani A, Harmanli O, Floro C, Gaughan JP, Hernandez E. Risk factors for obstetrical anal sphincter lacerations. Int Urogynecol J Pelvic Floor Dysfunct 2005;16:304–7.
3. Abbott D, Atere-Roberts N, Williams A, Oteng-Ntim E, Chappell LC. Obstetric anal sphincter injury [published erratum appears in BMJ 2010;341:c4256]. BMJ 2010;341:c3414.
4. 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.
5. Landy HJ, Laughon SK, Bailit J, Kominiarek MA, Gonzalez-Quintero VH, Ramirez M, et al. Characteristics associated with severe perineal and cervical lacerations during vaginal delivery. Obstet Gynecol 2011;117:627–35.
6. Gurol-Urganci I, Cromwell D, Edozien L, Mahmood TA, Adams EJ, Richmond DH, et al. Third-and fourth-degree perineal tears among primiparous women in England between 2000 and 2012: time trends and risk factors. BJOG 2013;120:1516–25.
7. Eason E, Labrecque M, Wells G, Feldman P. Preventing perineal trauma during childbirth: a systematic review. Obstet Gynecol 2000;95:464–71.
8. Johanson R, Pusey J, Livera N, Jones P. North Staffordshire/Wigan assisted delivery trial. Br J Obstet Gynaecol 1989;96:537–44.
9. Robinson JN, Norwitz ER, Cohen AP, McElrath TF, Lieberman ES. Episiotomy, operative vaginal delivery, and significant perineal trauma in nulliparous women. Am J Obstet Gynecol 1999;181:1180–4.
10. O'Mahony F, Hofmeyr GJ, Menon V. Choice of instruments for assisted vaginal delivery. The Cochrane Database of Systematic Reviews 2010, Issue 11. Art No.: CD005455. DOI: 10.1002/14651858.CD005455.pub2.
11. Gupta N, Dragovic K, Trester R, Blankstein J. The changing scenario of obstetrics and gynecology residency training. J Grad Med Educ 2015;7:401–6.
12. Safe prevention of the primary cesarean delivery. Obstetric Care Consensus No. 1. American College of Obstetricians and Gynecologists. Obstet Gynecol 2014;123:693–711.
13. Laughon SK, Berghella V, Reddy UM, Sundaram R, Lu Z, Hoffman MK. Neonatal and maternal outcomes with prolonged second stage of labor. Obstet Gynecol 2014;124:57–67.
14. Meister MR, Cahill AG, Conner SN, Woolfolk CL, Lowder JL. Predicting obstetric anal sphincter injuries in a modern obstetric population. Am J Obstet Gynecol 2016;215:310.e1–7.
15. Prevention and management of obstetric lacerations at vaginal delivery. Practice Bulletin No. 165. American College of Obstetricians and Gynecologists. Obstet Gynecol 2016;128:e1–15.
16. Handa VL, Danielsen BH, Gilbert WM. Obstetric anal sphincter lacerations. Obstet Gynecol 2001;98:225–30.
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. Baghestan E, Irgens LM, Børdahl PE, Rasmussen S. Trends in risk factors for obstetric anal sphincter injuries in Norway. Obstet Gynecol 2010;116:25–34.
19. Fenner DE, Genberg B, Brahma P, Marek L, DeLancey JO. Fecal and urinary incontinence after vaginal delivery with anal sphincter disruption in an obstetrics unit in the United States. Am J Obstet Gynecol 2003;189:1543–9.
20. Moraloglu O, Kansu-Celik H, Tasci Y, Karakaya BK, Yilmaz Y, Cakir E, et al. The influence of different maternal pushing positions on birth outcomes at the second stage of labor in nulliparous women. J Matern Fetal Neonatal Med 2017;30:245–9.
21. Le Ray C, Deneux-Tharaux C, Khireddine I, Dreyfus M, Vardon D, Goffinet F. Manual rotation to decrease operative delivery in posterior or transverse positions. Obstet Gynecol 2013;122:634–40.
22. California Health Care Foundation. Maternity care in California: delivering the data. Oakland (CA): California Health Care Foundation; 2016.