ORIGINAL ARTICLE
Subsequent Pregnancy Outcomes After Obstetric Anal Sphincter Injuries (OASIS) Elizabeth Basham, MD, Laura Stock, MD, Christina Lewicky-Gaupp, MD, Christopher Mitchell, BA, and Dana R. Gossett, MD, MSCI
Objectives: To describe obstetric outcomes in women with a prior obstetric anal sphincter injury (OASIS) and to identify risk factors for recurrence. Methods: A retrospective chart review of women who sustained an OASIS between November 2005 and March 2010 at a tertiary care hospital was performed to identify risk factors for recurrence. Results: One thousand six hundred twenty-nine patients had an OASIS. Of these, 758 patients (90%) subsequently delivered during the aforementioned timeframe; 685 patients had a subsequent vaginal delivery. Of the women, 3.2% had a recurrent OASIS. Recurrence was associated with larger birth weight (27% Q4000 g vs 11.6% G4000 g; P = 0.04) and delivery mode (25.0%, 12.5%, and 2.7% for forceps-assisted, vacuum-assisted, and spontaneous deliveries, respectively (P = 0.0001)), whereas a history of fourth-degree laceration, prior wound complication, or episiotomy at subsequent delivery were not (P = 0.5, P = 0.5, and P = 0.4, respectively). Conclusions: Recurrent OASIS occurred in a small percentage of women (3.2%) who subsequently delivered vaginally. Recurrent OASIS was associated with operative vaginal delivery and birth weight 4000 g or greater. Neither episiotomy at first delivery nor at subsequent delivery conferred an increased recurrence risk. Key Words: obstetric anal sphincter injury, operative delivery (Female Pelvic Med Reconstr Surg 2013;19: 328Y332)
T
he reported rate of obstetric anal sphincter injuries (OASIS) at the time of vaginal delivery varies from 0.0% to 23.9%, with the highest and lowest rates represented by studies with smaller sample sizes.1 Established risk factors for obstetrically related anal sphincter injury include primiparity; midline episiotomy; forceps delivery; prolonged second stage of labor; and increasing fetal weight, fetal position, and race.2Y7 These severe lacerations can result in significant morbidity, including chronic pain, infection, incontinence, embarrassment, and loss of sexual function.8Y11 There is a growing body of literature to support changes in obstetric practice to prevent OASIS during birth, including a decrease in use of episiotomies and operative vaginal deliveries, specifically the use of forceps.12Y14 In contrast, there are few studies looking at the risk of recurrence of OASIS in a subsequent pregnancy. Dandolu et al15 found an overall 5.76% recurrence rate in a subsequent pregnancy and an even higher rate in those with a fourth-degree laceration in their first pregnancy (7.73%). Payne et al16 reported a 10.7% rate of recurrent lacerations among 178 From the Department of Obstetrics and Gynecology, Feinberg School of Medicine, Northwestern University, Chicago, IL. Reprints: Christina Lewicky-Gaupp, MD, 676 N St. Claire, Suite 950, Chicago, IL 60611. E-mail: [email protected]. Reprints will not be available. The authors have declared they have no conflicts of interest. Copyright * 2013 by Lippincott Williams & Wilkins DOI: 10.1097/SPV.0b013e3182a5f98e
328
www.fpmrs.net
women with prior sphincter tears, while Harkin et al17 reported recurrent lacerations in 2 (4.4%) of 45 women. In a much larger study, Peleg et al18 reported recurrent lacerations in 58 (7.5%) of 774 women. As most of these studies have shown only modest increases in risk for recurrent OASIS, few risk factors for these recurrences have been identified and recommendations regarding delivery mode for a subsequent pregnancy are varied. The objective of this study was to determine the rate of recurrent OASIS in a patient population at a tertiary care hospital and to identify risk factors for recurrence; with this knowledge, we may be able to better identify, counsel, and modify labor practices in women at risk for this delivery complication.
MATERIALS AND METHODS This study was approved by the Northwestern University Feinberg School of Medicine Institutional Review Board. This was a planned secondary analysis of a large cohort of women who experienced a third- or fourth-degree perineal laceration at Prentice Women’s Hospital between November 2, 2005 and March 1, 2010.19 Patients were identified using the Northwestern Medical Enterprise Data Warehouse (EDW), an electronic repository of all data from the electronic medical records of Northwestern Memorial Hospital. Any data missing from that which was hard coded in the EDW was collected via chart review by the authors. Inclusion criteria for creation of the original cohort included an OASIS sustained during the aforementioned time period. Any tear that affected the anal sphincter muscles was classified as a third-degree laceration; and if the rectal mucosa was torn, the tear was identified as a fourthdegree laceration. Attending providers are required to check a box for either type of tear, and this information is then transferred into the EDW. Twin pregnancies were excluded. Data collected about the index pregnancy included demographic, medical, and obstetric variables, including birth weight, method of vaginal delivery (spontaneous or operative), indication for operative delivery, performance of episiotomy, and type and degree of tear (third or fourth). Similarly, postpartum wound complications were collected.19 For the current study, all patients who had a documented subsequent pregnancy and delivery at our institution during the study period were further analyzed. Data on the subsequent delivery were collected, including route of delivery, indication for any cesarean delivery, degree of subsequent perineal laceration, birth weight, and use of episiotomy. Statistical Package for the Social Sciences (version 18.0; SPSS Inc, Chicago, IL) was used for all data analyses. Categorical variables were evaluated using a W2 analysis, and continuous variables were evaluated using t tests. Binomial logistic regression was used to create several multivariable models incorporating those factors found to be significant in univariate analysis for prediction of recurrent severe lacerations. Alpha was set to 0.05 for all analyses.
Female Pelvic Medicine & Reconstructive Surgery
&
Volume 19, Number 6, November/December 2013
Copyright © 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Female Pelvic Medicine & Reconstructive Surgery
&
Volume 19, Number 6, November/December 2013 Pregnancy Outcomes After OASIS
RESULTS In the initial cohort, 1657 patients were identified by the EDW who met inclusion criteria of an OASIS during their index delivery. Of these, 28 subjects were excluded owing to twin gestations, for an ultimate sample size of 1629. Seven hundred fifty-eight patients (46%) went on to have a subsequent pregnancy in the aforementioned time period; 871 patients did not have documentation of a subsequent pregnancy at our institution (Fig. 1). Of the 758 women who had a documented subsequent delivery, 685 patients (90.4%) underwent a vaginal delivery; of these, 665 patients (87.7%) had spontaneous delivery, whereas 12 patients (1.6%) and 8 patients (1.1%) required forceps and vacuum assistance, respectively. Of the women, 9.6% had cesarean delivery. Of the women who had vaginal delivery, 627 women (84.5%) sustained a first- or secondary-degree tear and 23 women (3.2%) sustained a recurrent OASIS (21 women had a third-degree laceration and only 1 woman had a fourth-degree laceration). Demographic data are presented in Table 1. There were no differences in age, race, body mass index (BMI), or smoking status between the cohorts with and without subsequent OASIS. Women who experienced a recurrent OASIS were of higher parity than women who did not experience a recurrence. Risk factors for recurrent OASIS are presented in Table 2. The method of subsequent vaginal delivery was a significant predictor of recurrence, and recurrence was most common if operative vaginal delivery was used. Similarly, the mean T SD birth weight at subsequent delivery was found to be significantly higher in women who sustained a recurrent OASIS (3790.0 T 457.0 g vs 3486.7 T 483.2 g; P = 0.004), and macrosomia (Q4000 g) in the subsequent pregnancy conferred an increased risk of recurrence (Table 2). The mean T SD birth weight of subsequent babies in women who did not sustain a recurrent severe tear was significantly less than their previous child (j204.4 T 425.4 g vs +62.3 T 495.1 g; P = 0.01). Episiotomy in the subsequent delivery was not associated with recurrence. In the index delivery, episiotomy and use of chromic only suture for the primary repair were associated with recurrent OASIS in univariate analyses (Table 2), whereas higher-order wounds (ie, fourth-degree lacerations) and wound complications after index delivery did not confer an increased risk.
FIGURE 1. Patient selection. * 2013 Lippincott Williams & Wilkins
TABLE 1. Study Population
Age, mean T SD, yrs Race, n (%) White African American Asian Hispanic Hawaiian/ Pacific Islander Declined Smoker No Yes BMI, mean T SD Parity at subsequent delivery 1 2 3
No Recurrent OASIS (n = 663)
Recurrent OASIS (n = 22)
P*
31.5 T 3.9
32.3 T 3.4
0.9
487 (73.5) 22 (3.3)
20 (90.9) 0
36 (5.4) 45 (6.8) 20 (3.0)
0 1 (1.5) 0
53 (8.0)
1 (1.5)
646 (98.6) 9 (1.4) 28.6 T 4.2
22 (100) 0 27.9 T 3.8
630 (95.0) 27 (4.1) 6 (0.9)
16 (72.7) 5 (22.7) 6 (0.9)
0.6
0.6
0.9 G0.001
*Student t test (continuous variables) or W2 (categorical variables).
In multivariable analyses, both subsequent route of delivery and suture used at the initial repair remained significant predictors of recurrent OASIS, although episiotomy at first delivery did not. Three models were built using the following: (1) subsequent birth weight as a continuous variable, (2) differences in birth weight between the index and subsequent delivery as a continuous variable, and (3) macrosomia (defined as Q4000 g) as a dichotomous variable. In all 3 models, birth weight remained a significant predictor of recurrent OASIS, regardless of how it was assessed. Approximately 32% (23/73) of the cesarean deliveries performed in this cohort were primary elective sections. Indications for this mode of delivery were varied. One patient had suspected macrosomia, whereas 3 patients (13%) had a history of a shoulder dystocia. By far, the most common was a history of prior OASIS in 82.6%, often with documented complications including fecal incontinence in 4 women (17.4%), chronic pain and dyspareunia in 3 women, pelvic floor dysfunction in 3 women, prior wound breakdown in 3 women, and unspecified complications in 2 women. Women who had a fourth-degree laceration in their first pregnancy were more likely to have an elective cesarean delivery in their subsequent pregnancy compared to those who had a third-degree laceration (15.5% vs 8.7%; P = 0.029). Women who had a wound complication after their index OASIS (wound infection, breakdown, need for antibiotics, or further surgical debridement or repair) were not more likely to undergo cesarean delivery in the subsequent pregnancy (17.4% vs 10%’ P = 0.254); however, those women with wound complications were half as likely to have a subsequent pregnancy during the study period compared to those who did not have complications (35.4% vs 70.2%; P = 0.0001). The women who went on to have a subsequent documented delivery at our institution were of similar age to those who did not have another delivery (31.5 T 3.9 vs 31.6 T 5.4 years; P = 0.7). www.fpmrs.net
329
Copyright © 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Female Pelvic Medicine & Reconstructive Surgery
Basham et al
&
Volume 19, Number 6, November/December 2013
TABLE 2. Risk Factors for Subsequent OASIS Recurrent OASIS n (%); Mean T SD Method of vaginal delivery Spontaneous Forceps assisted Vacuum assisted Episiotomy at index delivery None Midline Mediolateral Suture for repair of index OASIS† Vicryl Vicryl + chromic Chromic Birth weight of subsequent delivery, mean T SD, g Difference between index and subsequent birth weights, g Subsequent birth weight, g * G4000 Q4000 Perineal tear in index delivery 3rd degree 4th degree Episiotomy at subsequent delivery None Midline Complications of index OASIS‡ None Documented complications
No Recurrent OASIS n (%); Mean T SD
P
RR (95% CI)
G0.0001 18 (2.7) 3 (25) 1 (12.5)
647 (97.3) 9 (75) 7 (87.5)
1.0 12.0 (3.0Y48.0) 5.1 (0.6Y44.0)
14 (2.4) 5 (3.2) 3 (15.8)
567 (97.6) 154 (96.8) 16 (84.2)
13 (2.6) 1 (1.1) 2 (14.3) 3790.0 T 457.0
469 (97.4) 94 (98.9) 12 (85.7) 3486.7 T 483.2
0.004
+62.3 T 495.1
j204.4 T 425.4
0.01
0.003 1.0 1.3 (0.5Y3.7) 7.3 (1.9Y28.0) 0.02 1.0 0.4 (0.1Y3.0) 7.4 (1.5Y37.8)
0.04 16 (2.4) 6 (6.3)
639 (97.6) 90 (93.7)
19 (2.9) 3 (2.9)
636 (97.1) 100 (97.1)
22 (3.0) 0 (0.0)
710 (97.0) 26 (100.0)
17 (2.9) 0 (0.0)
571 (97.1) 23 (100.0)
1.0 2.9 (1.1Y7.5) 0.9 1.0 1.1 (0.3Y3.8) 0.4 1.0 0.96 (0.95Y0.98) 0.4 1.0 0.96 (0.95Y0.98)
*Missing birth weight data for 2 patients. †Total number with documented suture type, 546; missing data, 139. ‡Total number with follow-up data, 548; missing data, 137.
The 2 cohorts were also of similar BMI (28.7 T 4.2 vs 28.8 T 4.5; P = 0.4), and the percentage of smokers was similar (1.9% vs 2.3%; P = 0.2). More parous women were less likely to have a documented subsequent delivery during the study time period; 94.6% of women who had a subsequent pregnancy were nulliparous vs 86.6% of those who did not (P G 0.001). In addition, white women were more likely to have a documented subsequent delivery than Black, Asian, Hispanic, American Indian, and Pacific Islander women (P G 0.001). Whereas diabetes alone was not a predictor of subsequent likelihood of pregnancy (P = 0.7), women with gestational diabetes in their index pregnancy were less likely to have a documented subsequent pregnancy (P = 0.002).
DISCUSSION Most (90%) of the women in this large retrospective study who had an OASIS at the time of their index pregnancy were subsequently delivered vaginally; the observed rate of recurrent sphincter injury in this study was low (3.2%) and compares favorably with other reported series. Recurrent OASIS was associated with operative vaginal delivery and birth weights of 4000 g or greater. On multivariable analysis, neither a history of an episiotomy nor an episiotomy in the subsequent delivery conferred an increased recurrence risk. Although a history of a
330
www.fpmrs.net
fourth-degree laceration and wound complication did not increase the risk of recurrence, women who had a wound complication with their index pregnancy were half as likely to have a subsequent pregnancy during the study period compared to those who did not have a previous wound complication. Those women whose index pregnancy fell later in our study period were, as one would expect, less likely to have a subsequent pregnancy during the follow-up period. Our rate of OASIS recurrence is favorably low and mostly comparable to other studies.2,6 For example, Harkin et al17 reported an OASIS recurrence rate of 4.4% in a cohort of more than 20,000 women. Our rate of recurrence is reassuring, as most women who had a subsequent vaginal delivery did so in an uncomplicated manner. In our study, episiotomy (neither at the time of index nor subsequent delivery) did not increase the risk of OASIS recurrence. This is in contrast to most other studies; in a study by Peleg et al,18 the risk of OASIS recurrence if midline episiotomy was performed in the subsequent delivery was 11% versus 2.1% if no episiotomy was performed. This was also true in a study by Payne et al,16 in which midline episiotomies are routinely performed; the recurrence rate of OASIS was 11%. In this cohort, the rate of episiotomy was lower than in most studies; and this may explain these findings. * 2013 Lippincott Williams & Wilkins
Copyright © 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Female Pelvic Medicine & Reconstructive Surgery
&
Volume 19, Number 6, November/December 2013 Pregnancy Outcomes After OASIS
Macrosomia is a known risk factor for OASIS,2,13 and our findings support this, as the mean birth weight of infants born to women who had a recurrent OASIS was significantly larger than the cohort of women who did not have recurrence, but not macrosomic based on usual obstetric definitions. However, these findings are consistent with other published studies. For example, Lowder et al20 found that infant birth weight of 3500 g or greater was associated with recurrent OASIS (odds ratio, 1.7; 95% confidence interval [CI], 1.1Y2.7). Similarly, Spydslaug et al21 reported that the absolute risks for recurrent anal sphincter laceration were 1.3% (95% CI, 0.4Y3.2) for birth weights less than 3000 g, and 23.3% (98% CI, 11.8Y38.6) for birth weights greater than 5000 g. Given the significantly increased risk found of recurrent OASIS in those women who delivered a larger infant, providers may wish to consider avoiding an operative delivery or offering these women an elective cesarean delivery in subsequent pregnancies. In this study, we found that the use of chromic suture use in the repair of the index laceration in patients increased the risk of a recurrent OASIS; and this finding has not been previously described. As this is based on such a small number of patients (and thus potentially limited by A-error), it is not possible to conclude that the use of chromic sutures is a risk factor for recurrent OASIS. However, it seems that most likely, the explanation for this is the decreased tensile strength of the scar when chromic is used; in a 2000 Cochrane Database review, an increased rate of suture dehiscence and resuturing of the perineal wound was reported when chromic catgut suture was used for obstetrical perineal laceration.22 Wound complications that occurred after the index laceration did not further increase risk for recurrent laceration, and this finding is novel. It suggests that despite having a complication secondary to OASIS, appropriate tissue healing occurred without any long-term sequelae. Although this is suggested by Baghestan et al,23 who showed no difference in pregnancy rates after a severe tear, that study did not address women with wound complications after their first delivery. Despite not incurring an increased risk of OASIS recurrence, it is interesting that these women were less likely to have a documented second pregnancy at our institution. This finding may suggest that these lacerations, and the complications that follow, are traumatic and result in avoidance of future pregnancy. Again, however, it should be noted that it is possible that some of these women did go on to have a subsequent delivery at another institution and were lost to our follow-up. Still, there is a dearth of literature examining the psychological impact of perineal trauma and future pregnancy, and further studies are needed. Women who previously had a fourth-degree laceration were more likely to be offered an elective cesarean delivery, although this group was no more likely to have a recurrent sphincter injury. Although the reason behind this in not clear, in general, women who had a fourth-degree laceration are more likely to have significant symptoms of anal incontinence compared to women who had a third-degree laceration; and it is possible that the presence or severity of these symptoms was paramount in the decision to deliver these women by elective cesarean delivery. Whereas literature regarding counseling of women who have persistent symptoms after their index tear is sparse, general expert opinion suggests offering women a cesarean delivery if they are symptomatic.24 In our study, this opinion was upheld, as most of the women with a prior fourth-degree tear who were offered a cesarean delivery were symptomatic from the index tear. As our provider base is very variable (with private faculty, academic faculty, and community health program faculty), however, it is not possible to determine * 2013 Lippincott Williams & Wilkins
how women with a history of OASIS are counseled regarding subsequent delivery. There are several limitations to this study. As we only had access to delivery data for subsequent pregnancies that occurred at our institution, the findings of the study may be slightly biased; women who sustained an OASIS may have subsequently delivered at another institution. However, our study population is still quite large; and thus, this concern may be offset somewhat by the population size. Additionally, we did not reabstract demographic data that might have changed since the index pregnancy as noted in Stock et al,19 including new medical morbidities, BMI, and age at second delivery. These factors may have affected the risk of recurrent OASIS; however, as they did not predict initial risk of laceration in the parent study, this seems unlikely. Finally, some of the subset analyses had very small numbers of subjects, which may have limited our ability to assess true differences between groups. In conclusion, we have found several significant risk factors for recurrent OASIS, including operative vaginal delivery and larger birth weight; episiotomy was not a risk factor. Overall, however, the risk of recurrent OASIS was relatively low, suggesting that most women with a history of an OASIS can safely undergo another vaginal delivery.
REFERENCES 1. Hordnes K, Bergsjo P. Severe lacerations after childbirth. Acta Obstet Gynecol Scand 1993;12:413Y422. 2. Landy HJ, Laughon SK, Bailit JL, et al. Characteristics associated with severe perineal and cervical lacerations during vaginal delivery. Obstet Gynecol 2011;117(3):627Y635. 3. Green JR, Soohoo SL. Factors associated with rectal injury in spontaneous deliveries. Obstet Gynecol 1989;73:732Y738. 4. Buchhave P, Flatow L, Rydhstroem H, et al. Risk factors for rupture of the anal sphincter. Eur J Obstet Gynecol Reprod Biol 1999;87:129Y132. 5. Zetterstro¨m J, Lo`pez A, Anze`n B, et al. Anal sphincter tears at vaginal delivery: risk factors and clinical outcome of primary repair. Obstet Gynecol 1999;94:21Y28. 6. Wood J, Amos L, Rieger N. Third degree anal sphincter tears: risk factors and outcome. Aust N Z J Obstet Gynaecol 1998;38(3):414Y417. 7. Sene`cal J, Xiong X, Fraser WD. Effect of fetal position on second-stage duration and labor outcome. Obstet Gynecol 2005;105:763Y772. 8. Samarasakera DN, Bekhit MT, Wright Y, et al. Long-term anal continence and quality of life following postpartum anal sphincter injury. Colorectal Dis 2008;10:793Y799. 9. Sorenson SM, Bondesen H, lstre O, et al. Perineal rupture following vaginal delivery; long-term consequences. Acta Obstet Gynecol Scand 1988;67:315Y318. 10. Goldaber K, Wendel P, McIntire D, et al. Postpartum perineal morbidity after fourth-degree perineal repair. Am J Obstet Gynecol 1993;168(2): 489Y493. 11. Kammerer-Doak DN, Wesol AB, Rogers RG, et al. A prospective cohort study of women after primary repair of obstetric anal sphincter laceration. Am J Obstet Gynecol 1999;181:1317Y1323. 12. Albers LL, Borders N. Minimizing genital tract trauma and related pain following spontaneous vaginal birth. J Midwifery Womens Health 2007;52(3):246Y253. 13. Fenner DE, Genberg B, Brahma P, et al. 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(6):1543Y1550.
www.fpmrs.net
331
Copyright © 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Basham et al
Female Pelvic Medicine & Reconstructive Surgery
14. Handa VL, Blomquist JL, McDermott KC, et al. Pelvic floor disorders after vaginal birth: effect of episiotomy, perineal laceration, and operative birth. Obstet Gynecol 2012;119(2 Pt 1):233Y239. 15. Dandolu V, Gaughan JP, Chatwani AJ, et al. Risk of recurrence of anal sphincter lacerations. Obstet Gynecol 2005;105:831Y835.
&
Volume 19, Number 6, November/December 2013
20. Lowder JL, Burrows LJ, Krohn MA, et al. Risk factors for primary and subsequent anal sphincter lacerations: a comparison of cohort by parity and prior mode of delivery. Am J Obstet Gynecol 2007;196(4): 344.e1Y344.e5.
16. Payne TN, Carey JC, Rayburn WF. Prior third- or fourth-degree perineal tears and recurrence risks. Int J Gynaecol Obstet 1999;64:55Y57.
21. Spydslaug A, Trogstad LI, Skrondal A, et al. Recurrent risk of anal sphincter laceration among women with vaginal deliveries. Obstet Gynecol 2005;105(2):307Y313.
17. Harkin R, Fitzpatrick M, O’Connell PR, et al. Anal sphincter disruption at vaginal delivery: is recurrence predictable? Eur J Obstet Gynecol Reprod Biol 2003;109:149Y152.
22. Kettle C, Johanson RB. Absorbable synthetic versus catgut suture material for perineal repair. Cochrane Database Syst Rev 2000;2:CD000006.
18. Peleg D, Kennedy CM, Merrill D, et al. Risk of repetition of a severe perineal laceration. Obstet Gynecol 1999;93:1021Y1024.
23. Baghestan E, Irgens LM, Børdahl PE, et al. Risk of recurrence and subsequent delivery after obstetric anal sphincter injuries. BJOG 2012;119(1);62Y69.
19. Stock L, Basham E, Gossett DR, et al. Factors associated with wound complications in women with obstetric anal sphincter injuries (OASIS). Am J Obstet Gynecol 2013;208(4):327.e1Y327.e6.
332
www.fpmrs.net
24. Sultan AH, Thakar RA, Fenner DE, eds. Perineal and Anal Sphincter Trauma. London, UK: Springer-Verlag; 2007.
* 2013 Lippincott Williams & Wilkins
Copyright © 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Subsequent Pregnancy Outcomes After Obstetric Anal Sphincter Injuries (OASIS) Elizabeth Basham, MD, Laura Stock, MD, Christina Lewicky-Gaupp, MD, Christopher Mitchell, BA, and Dana R. Gossett, MD, MSCI
Objectives: To describe obstetric outcomes in women with a prior obstetric anal sphincter injury (OASIS) and to identify risk factors for recurrence. Methods: A retrospective chart review of women who sustained an OASIS between November 2005 and March 2010 at a tertiary care hospital was performed to identify risk factors for recurrence. Results: One thousand six hundred twenty-nine patients had an OASIS. Of these, 758 patients (90%) subsequently delivered during the aforementioned timeframe; 685 patients had a subsequent vaginal delivery. Of the women, 3.2% had a recurrent OASIS. Recurrence was associated with larger birth weight (27% Q4000 g vs 11.6% G4000 g; P = 0.04) and delivery mode (25.0%, 12.5%, and 2.7% for forceps-assisted, vacuum-assisted, and spontaneous deliveries, respectively (P = 0.0001)), whereas a history of fourth-degree laceration, prior wound complication, or episiotomy at subsequent delivery were not (P = 0.5, P = 0.5, and P = 0.4, respectively). Conclusions: Recurrent OASIS occurred in a small percentage of women (3.2%) who subsequently delivered vaginally. Recurrent OASIS was associated with operative vaginal delivery and birth weight 4000 g or greater. Neither episiotomy at first delivery nor at subsequent delivery conferred an increased recurrence risk. Key Words: obstetric anal sphincter injury, operative delivery (Female Pelvic Med Reconstr Surg 2013;19: 328Y332)
T
he reported rate of obstetric anal sphincter injuries (OASIS) at the time of vaginal delivery varies from 0.0% to 23.9%, with the highest and lowest rates represented by studies with smaller sample sizes.1 Established risk factors for obstetrically related anal sphincter injury include primiparity; midline episiotomy; forceps delivery; prolonged second stage of labor; and increasing fetal weight, fetal position, and race.2Y7 These severe lacerations can result in significant morbidity, including chronic pain, infection, incontinence, embarrassment, and loss of sexual function.8Y11 There is a growing body of literature to support changes in obstetric practice to prevent OASIS during birth, including a decrease in use of episiotomies and operative vaginal deliveries, specifically the use of forceps.12Y14 In contrast, there are few studies looking at the risk of recurrence of OASIS in a subsequent pregnancy. Dandolu et al15 found an overall 5.76% recurrence rate in a subsequent pregnancy and an even higher rate in those with a fourth-degree laceration in their first pregnancy (7.73%). Payne et al16 reported a 10.7% rate of recurrent lacerations among 178 From the Department of Obstetrics and Gynecology, Feinberg School of Medicine, Northwestern University, Chicago, IL. Reprints: Christina Lewicky-Gaupp, MD, 676 N St. Claire, Suite 950, Chicago, IL 60611. E-mail: [email protected]. Reprints will not be available. The authors have declared they have no conflicts of interest. Copyright * 2013 by Lippincott Williams & Wilkins DOI: 10.1097/SPV.0b013e3182a5f98e
328
www.fpmrs.net
women with prior sphincter tears, while Harkin et al17 reported recurrent lacerations in 2 (4.4%) of 45 women. In a much larger study, Peleg et al18 reported recurrent lacerations in 58 (7.5%) of 774 women. As most of these studies have shown only modest increases in risk for recurrent OASIS, few risk factors for these recurrences have been identified and recommendations regarding delivery mode for a subsequent pregnancy are varied. The objective of this study was to determine the rate of recurrent OASIS in a patient population at a tertiary care hospital and to identify risk factors for recurrence; with this knowledge, we may be able to better identify, counsel, and modify labor practices in women at risk for this delivery complication.
MATERIALS AND METHODS This study was approved by the Northwestern University Feinberg School of Medicine Institutional Review Board. This was a planned secondary analysis of a large cohort of women who experienced a third- or fourth-degree perineal laceration at Prentice Women’s Hospital between November 2, 2005 and March 1, 2010.19 Patients were identified using the Northwestern Medical Enterprise Data Warehouse (EDW), an electronic repository of all data from the electronic medical records of Northwestern Memorial Hospital. Any data missing from that which was hard coded in the EDW was collected via chart review by the authors. Inclusion criteria for creation of the original cohort included an OASIS sustained during the aforementioned time period. Any tear that affected the anal sphincter muscles was classified as a third-degree laceration; and if the rectal mucosa was torn, the tear was identified as a fourthdegree laceration. Attending providers are required to check a box for either type of tear, and this information is then transferred into the EDW. Twin pregnancies were excluded. Data collected about the index pregnancy included demographic, medical, and obstetric variables, including birth weight, method of vaginal delivery (spontaneous or operative), indication for operative delivery, performance of episiotomy, and type and degree of tear (third or fourth). Similarly, postpartum wound complications were collected.19 For the current study, all patients who had a documented subsequent pregnancy and delivery at our institution during the study period were further analyzed. Data on the subsequent delivery were collected, including route of delivery, indication for any cesarean delivery, degree of subsequent perineal laceration, birth weight, and use of episiotomy. Statistical Package for the Social Sciences (version 18.0; SPSS Inc, Chicago, IL) was used for all data analyses. Categorical variables were evaluated using a W2 analysis, and continuous variables were evaluated using t tests. Binomial logistic regression was used to create several multivariable models incorporating those factors found to be significant in univariate analysis for prediction of recurrent severe lacerations. Alpha was set to 0.05 for all analyses.
Female Pelvic Medicine & Reconstructive Surgery
&
Volume 19, Number 6, November/December 2013
Copyright © 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Female Pelvic Medicine & Reconstructive Surgery
&
Volume 19, Number 6, November/December 2013 Pregnancy Outcomes After OASIS
RESULTS In the initial cohort, 1657 patients were identified by the EDW who met inclusion criteria of an OASIS during their index delivery. Of these, 28 subjects were excluded owing to twin gestations, for an ultimate sample size of 1629. Seven hundred fifty-eight patients (46%) went on to have a subsequent pregnancy in the aforementioned time period; 871 patients did not have documentation of a subsequent pregnancy at our institution (Fig. 1). Of the 758 women who had a documented subsequent delivery, 685 patients (90.4%) underwent a vaginal delivery; of these, 665 patients (87.7%) had spontaneous delivery, whereas 12 patients (1.6%) and 8 patients (1.1%) required forceps and vacuum assistance, respectively. Of the women, 9.6% had cesarean delivery. Of the women who had vaginal delivery, 627 women (84.5%) sustained a first- or secondary-degree tear and 23 women (3.2%) sustained a recurrent OASIS (21 women had a third-degree laceration and only 1 woman had a fourth-degree laceration). Demographic data are presented in Table 1. There were no differences in age, race, body mass index (BMI), or smoking status between the cohorts with and without subsequent OASIS. Women who experienced a recurrent OASIS were of higher parity than women who did not experience a recurrence. Risk factors for recurrent OASIS are presented in Table 2. The method of subsequent vaginal delivery was a significant predictor of recurrence, and recurrence was most common if operative vaginal delivery was used. Similarly, the mean T SD birth weight at subsequent delivery was found to be significantly higher in women who sustained a recurrent OASIS (3790.0 T 457.0 g vs 3486.7 T 483.2 g; P = 0.004), and macrosomia (Q4000 g) in the subsequent pregnancy conferred an increased risk of recurrence (Table 2). The mean T SD birth weight of subsequent babies in women who did not sustain a recurrent severe tear was significantly less than their previous child (j204.4 T 425.4 g vs +62.3 T 495.1 g; P = 0.01). Episiotomy in the subsequent delivery was not associated with recurrence. In the index delivery, episiotomy and use of chromic only suture for the primary repair were associated with recurrent OASIS in univariate analyses (Table 2), whereas higher-order wounds (ie, fourth-degree lacerations) and wound complications after index delivery did not confer an increased risk.
FIGURE 1. Patient selection. * 2013 Lippincott Williams & Wilkins
TABLE 1. Study Population
Age, mean T SD, yrs Race, n (%) White African American Asian Hispanic Hawaiian/ Pacific Islander Declined Smoker No Yes BMI, mean T SD Parity at subsequent delivery 1 2 3
No Recurrent OASIS (n = 663)
Recurrent OASIS (n = 22)
P*
31.5 T 3.9
32.3 T 3.4
0.9
487 (73.5) 22 (3.3)
20 (90.9) 0
36 (5.4) 45 (6.8) 20 (3.0)
0 1 (1.5) 0
53 (8.0)
1 (1.5)
646 (98.6) 9 (1.4) 28.6 T 4.2
22 (100) 0 27.9 T 3.8
630 (95.0) 27 (4.1) 6 (0.9)
16 (72.7) 5 (22.7) 6 (0.9)
0.6
0.6
0.9 G0.001
*Student t test (continuous variables) or W2 (categorical variables).
In multivariable analyses, both subsequent route of delivery and suture used at the initial repair remained significant predictors of recurrent OASIS, although episiotomy at first delivery did not. Three models were built using the following: (1) subsequent birth weight as a continuous variable, (2) differences in birth weight between the index and subsequent delivery as a continuous variable, and (3) macrosomia (defined as Q4000 g) as a dichotomous variable. In all 3 models, birth weight remained a significant predictor of recurrent OASIS, regardless of how it was assessed. Approximately 32% (23/73) of the cesarean deliveries performed in this cohort were primary elective sections. Indications for this mode of delivery were varied. One patient had suspected macrosomia, whereas 3 patients (13%) had a history of a shoulder dystocia. By far, the most common was a history of prior OASIS in 82.6%, often with documented complications including fecal incontinence in 4 women (17.4%), chronic pain and dyspareunia in 3 women, pelvic floor dysfunction in 3 women, prior wound breakdown in 3 women, and unspecified complications in 2 women. Women who had a fourth-degree laceration in their first pregnancy were more likely to have an elective cesarean delivery in their subsequent pregnancy compared to those who had a third-degree laceration (15.5% vs 8.7%; P = 0.029). Women who had a wound complication after their index OASIS (wound infection, breakdown, need for antibiotics, or further surgical debridement or repair) were not more likely to undergo cesarean delivery in the subsequent pregnancy (17.4% vs 10%’ P = 0.254); however, those women with wound complications were half as likely to have a subsequent pregnancy during the study period compared to those who did not have complications (35.4% vs 70.2%; P = 0.0001). The women who went on to have a subsequent documented delivery at our institution were of similar age to those who did not have another delivery (31.5 T 3.9 vs 31.6 T 5.4 years; P = 0.7). www.fpmrs.net
329
Copyright © 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Female Pelvic Medicine & Reconstructive Surgery
Basham et al
&
Volume 19, Number 6, November/December 2013
TABLE 2. Risk Factors for Subsequent OASIS Recurrent OASIS n (%); Mean T SD Method of vaginal delivery Spontaneous Forceps assisted Vacuum assisted Episiotomy at index delivery None Midline Mediolateral Suture for repair of index OASIS† Vicryl Vicryl + chromic Chromic Birth weight of subsequent delivery, mean T SD, g Difference between index and subsequent birth weights, g Subsequent birth weight, g * G4000 Q4000 Perineal tear in index delivery 3rd degree 4th degree Episiotomy at subsequent delivery None Midline Complications of index OASIS‡ None Documented complications
No Recurrent OASIS n (%); Mean T SD
P
RR (95% CI)
G0.0001 18 (2.7) 3 (25) 1 (12.5)
647 (97.3) 9 (75) 7 (87.5)
1.0 12.0 (3.0Y48.0) 5.1 (0.6Y44.0)
14 (2.4) 5 (3.2) 3 (15.8)
567 (97.6) 154 (96.8) 16 (84.2)
13 (2.6) 1 (1.1) 2 (14.3) 3790.0 T 457.0
469 (97.4) 94 (98.9) 12 (85.7) 3486.7 T 483.2
0.004
+62.3 T 495.1
j204.4 T 425.4
0.01
0.003 1.0 1.3 (0.5Y3.7) 7.3 (1.9Y28.0) 0.02 1.0 0.4 (0.1Y3.0) 7.4 (1.5Y37.8)
0.04 16 (2.4) 6 (6.3)
639 (97.6) 90 (93.7)
19 (2.9) 3 (2.9)
636 (97.1) 100 (97.1)
22 (3.0) 0 (0.0)
710 (97.0) 26 (100.0)
17 (2.9) 0 (0.0)
571 (97.1) 23 (100.0)
1.0 2.9 (1.1Y7.5) 0.9 1.0 1.1 (0.3Y3.8) 0.4 1.0 0.96 (0.95Y0.98) 0.4 1.0 0.96 (0.95Y0.98)
*Missing birth weight data for 2 patients. †Total number with documented suture type, 546; missing data, 139. ‡Total number with follow-up data, 548; missing data, 137.
The 2 cohorts were also of similar BMI (28.7 T 4.2 vs 28.8 T 4.5; P = 0.4), and the percentage of smokers was similar (1.9% vs 2.3%; P = 0.2). More parous women were less likely to have a documented subsequent delivery during the study time period; 94.6% of women who had a subsequent pregnancy were nulliparous vs 86.6% of those who did not (P G 0.001). In addition, white women were more likely to have a documented subsequent delivery than Black, Asian, Hispanic, American Indian, and Pacific Islander women (P G 0.001). Whereas diabetes alone was not a predictor of subsequent likelihood of pregnancy (P = 0.7), women with gestational diabetes in their index pregnancy were less likely to have a documented subsequent pregnancy (P = 0.002).
DISCUSSION Most (90%) of the women in this large retrospective study who had an OASIS at the time of their index pregnancy were subsequently delivered vaginally; the observed rate of recurrent sphincter injury in this study was low (3.2%) and compares favorably with other reported series. Recurrent OASIS was associated with operative vaginal delivery and birth weights of 4000 g or greater. On multivariable analysis, neither a history of an episiotomy nor an episiotomy in the subsequent delivery conferred an increased recurrence risk. Although a history of a
330
www.fpmrs.net
fourth-degree laceration and wound complication did not increase the risk of recurrence, women who had a wound complication with their index pregnancy were half as likely to have a subsequent pregnancy during the study period compared to those who did not have a previous wound complication. Those women whose index pregnancy fell later in our study period were, as one would expect, less likely to have a subsequent pregnancy during the follow-up period. Our rate of OASIS recurrence is favorably low and mostly comparable to other studies.2,6 For example, Harkin et al17 reported an OASIS recurrence rate of 4.4% in a cohort of more than 20,000 women. Our rate of recurrence is reassuring, as most women who had a subsequent vaginal delivery did so in an uncomplicated manner. In our study, episiotomy (neither at the time of index nor subsequent delivery) did not increase the risk of OASIS recurrence. This is in contrast to most other studies; in a study by Peleg et al,18 the risk of OASIS recurrence if midline episiotomy was performed in the subsequent delivery was 11% versus 2.1% if no episiotomy was performed. This was also true in a study by Payne et al,16 in which midline episiotomies are routinely performed; the recurrence rate of OASIS was 11%. In this cohort, the rate of episiotomy was lower than in most studies; and this may explain these findings. * 2013 Lippincott Williams & Wilkins
Copyright © 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Female Pelvic Medicine & Reconstructive Surgery
&
Volume 19, Number 6, November/December 2013 Pregnancy Outcomes After OASIS
Macrosomia is a known risk factor for OASIS,2,13 and our findings support this, as the mean birth weight of infants born to women who had a recurrent OASIS was significantly larger than the cohort of women who did not have recurrence, but not macrosomic based on usual obstetric definitions. However, these findings are consistent with other published studies. For example, Lowder et al20 found that infant birth weight of 3500 g or greater was associated with recurrent OASIS (odds ratio, 1.7; 95% confidence interval [CI], 1.1Y2.7). Similarly, Spydslaug et al21 reported that the absolute risks for recurrent anal sphincter laceration were 1.3% (95% CI, 0.4Y3.2) for birth weights less than 3000 g, and 23.3% (98% CI, 11.8Y38.6) for birth weights greater than 5000 g. Given the significantly increased risk found of recurrent OASIS in those women who delivered a larger infant, providers may wish to consider avoiding an operative delivery or offering these women an elective cesarean delivery in subsequent pregnancies. In this study, we found that the use of chromic suture use in the repair of the index laceration in patients increased the risk of a recurrent OASIS; and this finding has not been previously described. As this is based on such a small number of patients (and thus potentially limited by A-error), it is not possible to conclude that the use of chromic sutures is a risk factor for recurrent OASIS. However, it seems that most likely, the explanation for this is the decreased tensile strength of the scar when chromic is used; in a 2000 Cochrane Database review, an increased rate of suture dehiscence and resuturing of the perineal wound was reported when chromic catgut suture was used for obstetrical perineal laceration.22 Wound complications that occurred after the index laceration did not further increase risk for recurrent laceration, and this finding is novel. It suggests that despite having a complication secondary to OASIS, appropriate tissue healing occurred without any long-term sequelae. Although this is suggested by Baghestan et al,23 who showed no difference in pregnancy rates after a severe tear, that study did not address women with wound complications after their first delivery. Despite not incurring an increased risk of OASIS recurrence, it is interesting that these women were less likely to have a documented second pregnancy at our institution. This finding may suggest that these lacerations, and the complications that follow, are traumatic and result in avoidance of future pregnancy. Again, however, it should be noted that it is possible that some of these women did go on to have a subsequent delivery at another institution and were lost to our follow-up. Still, there is a dearth of literature examining the psychological impact of perineal trauma and future pregnancy, and further studies are needed. Women who previously had a fourth-degree laceration were more likely to be offered an elective cesarean delivery, although this group was no more likely to have a recurrent sphincter injury. Although the reason behind this in not clear, in general, women who had a fourth-degree laceration are more likely to have significant symptoms of anal incontinence compared to women who had a third-degree laceration; and it is possible that the presence or severity of these symptoms was paramount in the decision to deliver these women by elective cesarean delivery. Whereas literature regarding counseling of women who have persistent symptoms after their index tear is sparse, general expert opinion suggests offering women a cesarean delivery if they are symptomatic.24 In our study, this opinion was upheld, as most of the women with a prior fourth-degree tear who were offered a cesarean delivery were symptomatic from the index tear. As our provider base is very variable (with private faculty, academic faculty, and community health program faculty), however, it is not possible to determine * 2013 Lippincott Williams & Wilkins
how women with a history of OASIS are counseled regarding subsequent delivery. There are several limitations to this study. As we only had access to delivery data for subsequent pregnancies that occurred at our institution, the findings of the study may be slightly biased; women who sustained an OASIS may have subsequently delivered at another institution. However, our study population is still quite large; and thus, this concern may be offset somewhat by the population size. Additionally, we did not reabstract demographic data that might have changed since the index pregnancy as noted in Stock et al,19 including new medical morbidities, BMI, and age at second delivery. These factors may have affected the risk of recurrent OASIS; however, as they did not predict initial risk of laceration in the parent study, this seems unlikely. Finally, some of the subset analyses had very small numbers of subjects, which may have limited our ability to assess true differences between groups. In conclusion, we have found several significant risk factors for recurrent OASIS, including operative vaginal delivery and larger birth weight; episiotomy was not a risk factor. Overall, however, the risk of recurrent OASIS was relatively low, suggesting that most women with a history of an OASIS can safely undergo another vaginal delivery.
REFERENCES 1. Hordnes K, Bergsjo P. Severe lacerations after childbirth. Acta Obstet Gynecol Scand 1993;12:413Y422. 2. Landy HJ, Laughon SK, Bailit JL, et al. Characteristics associated with severe perineal and cervical lacerations during vaginal delivery. Obstet Gynecol 2011;117(3):627Y635. 3. Green JR, Soohoo SL. Factors associated with rectal injury in spontaneous deliveries. Obstet Gynecol 1989;73:732Y738. 4. Buchhave P, Flatow L, Rydhstroem H, et al. Risk factors for rupture of the anal sphincter. Eur J Obstet Gynecol Reprod Biol 1999;87:129Y132. 5. Zetterstro¨m J, Lo`pez A, Anze`n B, et al. Anal sphincter tears at vaginal delivery: risk factors and clinical outcome of primary repair. Obstet Gynecol 1999;94:21Y28. 6. Wood J, Amos L, Rieger N. Third degree anal sphincter tears: risk factors and outcome. Aust N Z J Obstet Gynaecol 1998;38(3):414Y417. 7. Sene`cal J, Xiong X, Fraser WD. Effect of fetal position on second-stage duration and labor outcome. Obstet Gynecol 2005;105:763Y772. 8. Samarasakera DN, Bekhit MT, Wright Y, et al. Long-term anal continence and quality of life following postpartum anal sphincter injury. Colorectal Dis 2008;10:793Y799. 9. Sorenson SM, Bondesen H, lstre O, et al. Perineal rupture following vaginal delivery; long-term consequences. Acta Obstet Gynecol Scand 1988;67:315Y318. 10. Goldaber K, Wendel P, McIntire D, et al. Postpartum perineal morbidity after fourth-degree perineal repair. Am J Obstet Gynecol 1993;168(2): 489Y493. 11. Kammerer-Doak DN, Wesol AB, Rogers RG, et al. A prospective cohort study of women after primary repair of obstetric anal sphincter laceration. Am J Obstet Gynecol 1999;181:1317Y1323. 12. Albers LL, Borders N. Minimizing genital tract trauma and related pain following spontaneous vaginal birth. J Midwifery Womens Health 2007;52(3):246Y253. 13. Fenner DE, Genberg B, Brahma P, et al. 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(6):1543Y1550.
www.fpmrs.net
331
Copyright © 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Basham et al
Female Pelvic Medicine & Reconstructive Surgery
14. Handa VL, Blomquist JL, McDermott KC, et al. Pelvic floor disorders after vaginal birth: effect of episiotomy, perineal laceration, and operative birth. Obstet Gynecol 2012;119(2 Pt 1):233Y239. 15. Dandolu V, Gaughan JP, Chatwani AJ, et al. Risk of recurrence of anal sphincter lacerations. Obstet Gynecol 2005;105:831Y835.
&
Volume 19, Number 6, November/December 2013
20. Lowder JL, Burrows LJ, Krohn MA, et al. Risk factors for primary and subsequent anal sphincter lacerations: a comparison of cohort by parity and prior mode of delivery. Am J Obstet Gynecol 2007;196(4): 344.e1Y344.e5.
16. Payne TN, Carey JC, Rayburn WF. Prior third- or fourth-degree perineal tears and recurrence risks. Int J Gynaecol Obstet 1999;64:55Y57.
21. Spydslaug A, Trogstad LI, Skrondal A, et al. Recurrent risk of anal sphincter laceration among women with vaginal deliveries. Obstet Gynecol 2005;105(2):307Y313.
17. Harkin R, Fitzpatrick M, O’Connell PR, et al. Anal sphincter disruption at vaginal delivery: is recurrence predictable? Eur J Obstet Gynecol Reprod Biol 2003;109:149Y152.
22. Kettle C, Johanson RB. Absorbable synthetic versus catgut suture material for perineal repair. Cochrane Database Syst Rev 2000;2:CD000006.
18. Peleg D, Kennedy CM, Merrill D, et al. Risk of repetition of a severe perineal laceration. Obstet Gynecol 1999;93:1021Y1024.
23. Baghestan E, Irgens LM, Børdahl PE, et al. Risk of recurrence and subsequent delivery after obstetric anal sphincter injuries. BJOG 2012;119(1);62Y69.
19. Stock L, Basham E, Gossett DR, et al. Factors associated with wound complications in women with obstetric anal sphincter injuries (OASIS). Am J Obstet Gynecol 2013;208(4):327.e1Y327.e6.
332
www.fpmrs.net
24. Sultan AH, Thakar RA, Fenner DE, eds. Perineal and Anal Sphincter Trauma. London, UK: Springer-Verlag; 2007.
* 2013 Lippincott Williams & Wilkins
Copyright © 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
