Endoanal Ultrasound Findings and Fecal Incontinence Symptoms in Women With and Without Recognized Anal Sphincter Tears

Holly E. Richter, PhD, MD, Julia R. Fielding, MD, Catherine S. Bradley, MD, MSCE, Victoria L. Handa, MD, Paul Fine, MD, Mary Pat FitzGerald, MD, Anthony Visco, MD, Arnold Wald, MD, Christiane Hakim, MD, J. T. Wei, MD, and Anne M. Weber, MD, for the Pelvic Floor Disorders Network*
OBJECTIVE: To estimate whether endoanal ultrasound findings are more prevalent in primiparous women with a history of anal sphincter tear than in women without this history and whether the findings are associated with fecal incontinence symptoms. METHODS: A total of 251 primiparous women at seven clinical sites underwent standardized ultrasound assessment of the internal and external anal sphincter 6 12 months after delivery. Participants were women in the three cohorts of the Childbirth and Pelvic Symptoms Study: 1) women with clinically evident third- or fourthdegree tear at vaginal delivery (n 106); 2) no tear at vaginal delivery (n 106); and 3) cesarean delivery without labor (n 39). Women completed the Fecal Incontinence Severity Index to assess fecal incontinence symptoms. RESULTS: Thirty-five percent of the sphincter tear group exhibited internal sphincter gaps compared with 3% of vaginal controls (odds ratio [OR] 18.4, 95% confidence interval [CI] 5.5 62.1) and 10% of cesarean controls. External sphincter gaps were identified in 51% of the tear group compared with 31% of vaginal controls (OR 2.3, 95% CI 1.3 4.0) and 28% of cesarean controls. In the tear group, fecal incontinence severity was greater in those with internal sphincter gaps compared with those with no internal sphincter gaps (Fecal Incontinence Severity Index score 6.6 8.3 compared with 3.3 6.1, P .02), as well as in those with external sphincter gaps (6.1 8.4 compared with 2.7 5.0, P .01), and greatest in those with both internal and external sphincter gaps compared with at least one gap not present (7.2 8.1 compared with 3.4 6.4, P .003). CONCLUSION: Anal sphincter gaps detected by ultrasonography are prevalent in postpartum primiparous women with a history of sphincter tear and are associated with fecal incontinence severity.
(Obstet Gynecol 2006;108:1394401)

* For a listing of members of the Pelvic Floor Disorders Network, see the Appendix. From the Department of Obstetrics and Gynecology, University of Alabama at Birmingham, Birmingham, Alabama; Department of Radiology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina; Department of Obstetrics and Gynecology, University of Iowa Carver College of Medicine Iowa City, Iowa; Johns Hopkins School of Medicine, Baltimore, Maryland; Baylor College of Medicine, Houston, Texas; Department of Obstetrics and Gynecology, Loyola University Chicago, Chicago, Illinois; Department of Obstetrics and Gynecology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina; Department of Medicine, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania; Division of Radiology, Magee-Womens Hospital, Pittsburgh, Pennsylvania; Department of Urology, University of Michigan, Ann Arbor, Michigan; National Institute of Child Health and Human Development, Bethesda, Maryland. Supported by grants From the National Institute of Child Health and Human Development (U01 HD41249, U10 HD41268, U10 HD41248, U10 HD41250, U10 HD41261, U10 HD41263, U10 HD41269, and U10 HD41267) and the National Institute of Diabetes, Digestive and Kidney Diseases (K24 DK068389). The authors thank Dr. Robert Park, the Chair of the Pelvic Floor Disorders Network Steering Committee, for his contributions in planning the study. The authors gratefully acknowledge the other investigators who performed the imaging studies at all of the clinical sites, including Gregg Shore, MD, Franklin Tessler, MD, and Mark Lockhart, MD, University of Alabama at Birmingham; Bruce Brown, MD, Alan Stolpen, MD, PhD, University of Iowa; Susan Gearhart, MD, Harpreet Pannu, MD, Johns Hopkins Medical Institutes; and Caryl Solomon, MD, Loyola University. Presented as a poster at the Society of Gynecologic Investigators Meeting, Toronto, Ontario, Canada, March 24, 2005. Corresponding author: Holly E. Richter, PhD, MD, Medical Surgical Gynecology, University of Alabama at Birmingham, 620 20th Street South, NHB 219, Birmingham, AL 35249-7333; e-mail: hrichter@uabmc.edu. 2006 by The American College of Obstetricians and Gynecologists. Published by Lippincott Williams & Wilkins. ISSN: 0029-7844/06

LEVEL OF EVIDENCE: II-2

major cause of fecal incontinence in young healthy women is anal sphincter damage during childbirth,1 occurring in as many as 18% of vaginal deliveries2 in the United States. Within the first 6 months after repair of obstetric anal sphincter tears,

1394

VOL. 108, NO. 6, DECEMBER 2006

OBSTETRICS & GYNECOLOGY

29 53% of women report gas incontinence, and 510% report stool incontinence.3 6 Fecal incontinence may be related to persistent sphincter damage because up to 66% of primiparous women demonstrate a defect within the anal sphincter by endoanal ultrasonography 3 months after repair.3 In addition, occult sphincter tears (not apparent at delivery, but present on ultrasonogram) occur in 2335% of primiparous women at vaginal delivery.27 Endoanal ultrasonography has been the standard for radiologic evaluation of the anal sphincter. Although ultrasonography reliably identifies anatomic defects or thinning of the internal anal sphincter,8 interpretation of external anal sphincter images is more subjective, operator-dependent, and confounded by normal anatomic variations including an external sphincter gap in many normal patients.9,10 With these limitations, ultrasound findings of anal sphincter tear are highly correlated with electromyogram and manometric findings and are both sensitive and specific for anal sphincter tear at subsequent surgery.1114 Ultrasound findings may be related to the severity of fecal incontinence symptoms, but studies to date3,1518 have small numbers and short-term followup. Our purpose was to estimate whether primiparous women with sphincter tear at vaginal delivery have increased ultrasound findings of sphincter gaps and whether ultrasound findings 6 12 months after delivery are associated with fecal incontinence symptoms.

MATERIALS AND METHODS

The Childbirth and Pelvic Symptoms (CAPS) study evaluated fecal and urinary incontinence symptoms at 6 weeks and 6 months after first delivery and determined that fecal incontinence was more common in women after vaginal delivery with an anal sphincter tear, compared with those after vaginal delivery without a sphincter tear.6 This paper reports on the Childbirth and Pelvic SymptomsImaging Supplementary Study (CAPSIS) performed in a subset of the Childbirth and Pelvic Symptoms study population. Both studies were performed by the Pelvic Floor Disorders Network, a cooperative agreement network sponsored by the National Institute of Child Health and Human Development. All but three women in this study were recruited from 921 women enrolled in the Childbirth and Pelvic Symptoms study. The other three women fulfilled the eligibility criteria for this study but were not in the Childbirth and Pelvic Symptoms study; they also fulfilled the eligibility criteria of the Childbirth and Pelvic Symptoms study (except possibly for an English language eligibility requirement in the Childbirth and Pelvic Symptoms study). Methods of the Childbirth and

Pelvic Symptoms study have been reported in detail and are briefly summarized here.6 Women were enrolled into three groups of primiparous women identified while hospitalized after a singleton delivery: 1) 407 women delivered vaginally with a clinically evident anal sphincter tear (sphincter tear group), 2) 390 women delivered vaginally without evidence of an anal sphincter tear (vaginal control group), and 3) 124 women delivered by cesarean before labor (cesarean control group). Women with anal sphincter tears included those with third-degree perineal tears (into or through the anal sphincter) and those with fourth-degree tears (through the anal sphincter and rectal mucosa). Childbirth and Pelvic Symptoms study participants completed questionnaires about pelvic symptoms and quality of life by telephone at 6 weeks and 6 months postpartum. Women were invited to participate in this imaging study at or after their 6-month interviews for the Childbirth and Pelvic Symptoms study. Separate institutional review board approval and written informed consent for this study was obtained from each clinical site and data coordination center. All study participants completed the Fecal Incontinence Severity Index,19 which assesses the frequency of four symptoms (incontinence of gas, mucus, liquid stool, and solid stool) using the following scale: 2 or more times a day, once a day, 2 or more times a week, once a week, 13 times per month, or never. A higher score on the Fecal Incontinence Severity Index represents greater fecal incontinence severity, including flatus, with a potential range of 0 61. Prevalence of fecal incontinence was defined as any involuntary leakage of mucus, liquid stool, or solid stool on the Fecal Incontinence Severity Index (flatus was not included in the definition of fecal incontinence). The imaging studies were performed on all participants between 6 and 12 months postpartum after a negative urine pregnancy test. When imaging was performed more than 4 weeks after the final interview in the Childbirth and Pelvic Symptoms study (which included the Fecal Incontinence Severity Index), the Fecal Incontinence Severity Index was repeated at the time of the ultrasound examination. The ultrasound examination was performed and interpreted by physician investigators, all with significant clinical experience in ultrasonography of the anal sphincter. The investigators were blinded to cohort group and symptoms. To improve consistency in performing and interpreting the ultrasonogram, the radiologists and other experienced investigators performing the ultrasound examinations underwent centralized training with one radiologist trainer (J.R.F.), expert in pelvic imaging, before data collection. Ultrasound train-

VOL. 108, NO. 6, DECEMBER 2006

Richter et al

Fecal Incontinence Symptoms and Sphincter Gaps

1395

ing consisted of the review of normal and abnormal examinations, with specific emphasis on the definition of internal and external sphincter gaps. Each participant was studied in the left lateral decubitus position. An endoanal probe (B&K Medical, Sandtoften, Denmark; 10 MHz, 360 window) was placed into the rectum, and the most superior aspect of the anal sphincter was identified by visualization of the V-shaped puborectalis muscle. Contiguous images, 5 mm in thickness, extending from cephalad to caudad were recorded on paper or film. Videotaping was not performed because not all sites had this capability. Investigators were then free to examine the anal canal using continuous movement to better assess muscle integrity. The internal sphincter was identified as a concentric hypoechoic band surrounding the anal mucosa. The external sphincter was identified lateral to the internal sphincter as a concentric band of mixed echogenicity. An echogenic or echolucent disruption in the sphincter(s) seen on one or more images (each 5 mm in thickness) was defined as a gap. They were located by clock-face position to the nearest hour, with anterior midline as 12 oclock. To assess reliability approximately, 1 of 10 ultrasound cases (from sites other than the expert radiologist) were reviewed by the expert radiologist using contiguous static images.8,20 We hypothesized that women who delivered vaginally and required repair of clinically evident sphincter tears would have more sphincter gaps than women who delivered vaginally without overt sphincter tears. A sample size of 100 women with and 100 without anal sphincter tears after vaginal delivery provided 80% power to identify a 20% difference (eg, 40% compared with 60%) in sphincter gaps between the two groups using a two-sided 5% test of significance. We also hypothesized that women who delivered vaginally without clinically evident sphincter tears were more likely to have sphincter gaps on imaging than women delivered by cesarean before labor. However, because there were far fewer cesarean controls in the original study, this study was not powered to detect differences between the cesarean and vaginal control groups. The primary analyses compared the sphincter tear group with the vaginal control group. Continuous measures were compared by two-tailed t tests. Because of a lack of normality, the Fecal Incontinence Severity Index severity score was compared by the MannWhitney test. Dichotomous measures were compared by Fisher exact test. Models that adjust for site-to-site variability were not fitted because of the small number of gaps identified at each site. As a result of the small numbers, estimates of the severity

score on the Fecal Incontinence Severity Index by site and gap or no gap had large standard errors. Cohens was computed to assess interrater reliability. Data are presented as means and standard deviation (SD), counts and percentages, and odds ratios (OR) and 95% confidence intervals (CI).

RESULTS
A total of 256 women were enrolled in this study (109 women in the sphincter tear group, 108 in the vaginal control group, and 39 in the cesarean control group). Ultrasound examinations were not performed in five participants (three in the sphincter tear group and two in the vaginal control group). The women who participated were similar to the Childbirth and Pelvic Symptoms study women who did not participate in age, rates of episiotomy and of vacuum or forceps intervention, and fecal incontinence prevalence and severity. However, more African-American women participated in this study. Ultrasound examinations were not performed in five participants, leaving 106, 106, and 39 women in the three groups, respectively. Most women (92%) had their ultrasound examinations 6 9 months after delivery; the remaining ultrasound examinations were performed 10 12 months after delivery. Characteristics of the population are summarized in Table 1. Comparing the women with sphincter tears with the vaginal controls, we found no significant differences in age, marital status, race, or education. Mean infant birth weight was approximately 200 g lower in the vaginal control group compared with the sphincter tear group (P .001). Women in the vaginal control group were less likely to have had operative (forceps or vacuum) delivery (P .001). The frequency of sphincter gaps on ultrasound examination is described in Table 2. Gaps in the internal sphincter were identified in 35% (n 37) of the sphincter tear group, in 3% (n 3) of vaginal controls (P .001), and in 10% (n 4) of cesarean delivery controls. Therefore, the relationship between ultrasound gaps and fecal incontinence symptoms was performed in the tear group only. Within the sphincter tear group, more internal sphincter gaps were identified in women with fourth-degree tears than in those with third-degree tears (78% compared with 20%, P .001). The midpoint of the internal sphincter gap was in the anterior (10:00 to 2:00 oclock) segment of the sphincter in all but one woman in the sphincter tear group. In the sphincter tear group, 27% of women with an internal sphincter gap had symptoms of fecal incontinence, compared with 13% of those without an internal sphincter gap (OR 2.5, 95% CI 0.9 6.8). In

1396

Richter et al

Fecal Incontinence Symptoms and Sphincter Gaps

OBSTETRICS & GYNECOLOGY

Table 1. Characteristics of Study Sample by Cohort

Sphincter Tear (n 109)
27.5 6.2 67 (62.0) 79 (72.5) 25 (22.9) 5 (4.6) 5 (4.6) 22 (20.2) 30 (27.5) 32 (29.4) 20 (18.3) 3,596 438* 60 (56.1)* 45 (44.6)*

Variable
Age (y) Married Race White/Caucasian Black/African American Other Education Less than high school High school graduate-equivalent Some college College graduate Graduate degree Birth weight (g) Forceps or vacuum Midline episiotomy

Vaginal Control (n 108)

26.1 5.6 63 (59.4) 71 (65.7) 29 (26.9) 8 (7.4) 7 (6.5) 31 (28.7) 27 (25.0) 32 (29.6) 11 (10.2) 3,406 372* 17 (15.6)* 19 (18.3)*

Cesarean Delivery (n 39)

28.3 7.4 24 (61.5) 29 (74.4) 9 (23.1) 1 (2.6) 4 (10.3) 7 (17.9) 12 (30.8) 9 (23.1) 7 (17.9) 3,526 483 Not applicable Not applicable

Data are expressed as mean standard deviation or n (%). * Denotes a statistically significant (P .05) difference between the sphincter tear and vaginal control cohorts.

Table 2. Ultrasound Findings of Anal Sphincter Gaps

Sphincter Third Fourth Tear Degree Degree [n (%)] [n (%)] [n (%)] OR (95% CI)* Vaginal Control [n (%)] OR (95% CI) Cesarean Control [n (%)]

Study Cohort

n 106 79 27 106 39 Internal anal sphincter gap 37 (35) 16 (20) 21 (78) 13.8 (4.839.8) 3 (3) 18.4 (5.562.1) 4 (10) P .001 .001 External anal sphincter gap 54 (51) 34 (43) 20 (74) 3.8 (1.410.0) 33 (31) 2.3 (1.34.0) 11 (28) P .007 .005 Both internal and external anal sphincter gaps 29 (27) 11 (14) 18 (67) 12.4 (4.434.4) 2 (2) 19.6 (4.584.5) 2 (5) P .001 .001 25/69 (36) 23/63 (37) 2/6 (33) 0.9 (0.24.9) 31/103 (30) 1.3 (0.72.5) 9/35 (26) External anal sphincter gap only P .9 .41 19 (18) 12 (15) 7 (26) 2.0 (0.75.8) 5 (5) 4.4 (1.612.3) 5 (13) Fecal Incontinence P .25 .004
OR, odds ratio; CI, confidence interval. * Odds ratios and P values for comparisons between third- and fourth-degree tear groups. Odds ratios and P values for comparisons between sphincter tear and vaginal control groups. Study participants with an internal anal sphincter gap are excluded from both the numerator and denominator. Fecal incontinence was defined as any involuntary leakage of mucus, liquid stool, or solid stool on the Fecal Incontinence Severity Index.

addition, women with an internal sphincter gap had a mean Fecal Incontinence Severity Index score of 6.6 8.3, compared with 3.3 6.1 in those without an internal sphincter gap (P .008) (Table 3). The likelihood of an external sphincter gap was also greater in the sphincter tear group (51%) than in the vaginal control group (31%) (OR 2.3, 95% CI 1.3 4.0) or the cesarean control group (28%) (Table 2). As with internal sphincter gaps, external sphincter gaps in women with fourth-degree repairs were significantly more common than in those with thirddegree repairs (74% compared with 43%, OR 3.8, 95% CI 1.4 10.0). The midpoint of the external sphincter gap was in the anterior (10:00 to 2:00

oclock) segment of the sphincter in all but two (98%) women in the sphincter tear group. Within the sphincter tear group, women with an external sphincter gap were more likely to have symptoms of fecal incontinence than women without such a gap (OR 3.3, 95% CI 1.19.9) (Table 3). Subjects with an external sphincter gap (with or without an internal sphincter gap) had a mean Fecal Incontinence Severity Index score of 6.1 8.4, compared with 2.7 5.0 in those without an external sphincter gap (P .010). When considering only those with an isolated external sphincter gap (ie, excluding those with a concomitant internal sphincter gap), the prevalence of external

VOL. 108, NO. 6, DECEMBER 2006

Richter et al

Fecal Incontinence Symptoms and Sphincter Gaps

1397

Table 3. Relationship Between Ultrasound Findings of Sphincter Gaps and Fecal Incontinence Prevalence and Severity in the Sphincter Tear Group
Fecal Fecal Incontinence Severity Index Score Incontinence Mean SD (n) Median (IQR) [n (%)]
10/37 (27) 9/69 (13) .11 14/54 (26) 5/52 (10) .042 6/25 (24) 3/44 (7) .063 8/29 (28) 11/77 (14) .15 6.6 8.3 (37) 3.3 6.1 (69) .02 6.1 8.4 (54) 2.7 5.0 (52) .01 4.9 8.7 (25) 2.3 3.8 (44) .40 7.2 8.1 (29) 3.4 6.4 (77) .003 4 (011) 0 (04) 4 (08) 0 (04) 0 (08) 0 (04) 4 (011) 0 (04)

Ultrasound Findings
Internal anal sphincter gap No internal anal sphincter gap P External anal sphincter gap No external anal sphincter gap P External anal sphincter gap only* No external or internal anal sphincter gap P Concurrent internal and external anal sphincter gaps At least one internal and external anal sphincter gap is not present P

SD, standard deviation; IQR, interquartile range. * Study participants with an internal anal sphincter gap are excluded from the numerator.

sphincter gaps in the sphincter tear group and the vaginal control group was not significantly different (24% compared with 30%, respectively). In those without an internal sphincter gap, there was no difference between those with fourth- and third-degree tears (2 of 6, 33% compared with 24 of 64, 37%, respectively). The mean Fecal Incontinence Severity Index score in the 25 women with an isolated external sphincter gap in the sphincter tear group was 4.9 8.7, whereas the mean Fecal Incontinence Severity Index score without either tear was 2.3 3.8 (P .40) (Table 3). Symptoms were most severe among women in the sphincter tear group who had both internal and external anal sphincter gaps, with a mean Fecal Incontinence Severity Index score of 7.2 8.1, compared with women in the sphincter tear group without gaps in both the internal and external anal sphincters, 3.4 6.4 (P .003) (Table 3). Overall, more women in the tear group had fecal incontinence symptoms than did vaginal controls (18% compared with 5%, OR 4.4, CI 1.6 12.3) (Table 2). Twenty-six percent (7 of 27) of women with fourth-degree sphincter tears had fecal incontinence symptoms, compared with 15% (12 of 79) with thirddegree tear, although this difference did not achieve statistical significance (OR 2.0, 95% CI 0.75.6) (Table 2). Fecal incontinence symptoms were more severe among women sustaining a fourth-degree compared with a third-degree tear (Fecal Incontinence Severity Index score 7.5 10.1 compared with 3.4 5.4, respectively, P .06) although statistical significance was marginal. The expert radiologist trainer reviewed 19 of 206 ultrasound examinations conducted at the other clinical

sites to assess their reliability. Gaps in the internal sphincter were identified on the images of eight of 19 participants: four by both readers, one by the clinical site only, and three by the central reader only ( 0.52). Gaps in the external sphincter were identified on 10 of 19 images: four by both readers, four by the clinical site only, and two by the central reviewer only ( 0.33).

DISCUSSION
This study reports on the number of sphincter gaps seen by ultrasonography 6 months postpartum in a cohort of women sustaining an anal sphincter tear compared with those women without sphincter tears, and the association of gaps with bowel symptoms. Women who sustained clinically apparent anal sphincter damage at the time of delivery, especially fourth-degree tears, were more likely to have ultrasound findings of internal and external anal sphincter gaps despite primary repair, compared with women without apparent anal sphincter tear at delivery. This study also showed that sphincter gaps were related to the degree of tear at the time of delivery. In addition, in the tear group, gaps were associated with increased severity of fecal incontinence (significant) and a higher rate of fecal incontinence (which did not achieve statistical significance) compared with those women with tears but without a gap. When there were both external and internal sphincter gaps, it was likely to indicate a tear with increased severity and with a slightly increased likelihood of fecal incontinence. However, an external sphincter gap without a concomitant internal sphincter gap was prevalent in all the cohorts and was not associated with an increased severity of fecal incontinence. Our findings are consistent with prior reports of

1398

Richter et al

Fecal Incontinence Symptoms and Sphincter Gaps

OBSTETRICS & GYNECOLOGY

sonographic sphincter gaps and increased fecal incontinence symptoms among women sustaining fourth- compared with third-degree tears.2,21,22 In one previous study,21 women with fourth-degree tears were more likely to have combined internal and external sphincter gaps and had more internal sphincter gaps than women with third-degree tears. A significant correlation existed between bowel symptoms and persistent anal sphincter gaps. In our study, women with fourth-degree tears had more internal sphincter gaps (and more combined external and internal sphincter gaps) than those with third-degree tears. Although both prevalence and severity of fecal incontinence symptoms trended toward being significantly increased in the fourth-degree tear group, we did not have a large enough sample size to show this difference. In the larger Childbirth and Pelvic Symptoms study, women with fourth-degree tears (n 87) reported a higher prevalence of fecal incontinence (26% compared with 15%, P .02) and a higher fecal incontinence severity score (6.2 8.9 compared with 3.2 5.5, P .009) at 6 months, compared with women with third-degree tears (n 320).6 Other studies have used ultrasonography to evaluate women with a history of sphincter tear.2329 With evaluation times ranging from 6 weeks to 5 years after delivery, all found significant numbers of persistent sphincter gaps (notably in the internal anal sphincter) that were associated with increased bowel symptoms. None of these studies highlighted the increased risk of internal anal sphincter gaps after fourth-degree tears compared with third-degree tears. Gaps in the anal sphincters of women who underwent sphincter repair after delivery might result from poor healing or from an inadequate surgical repair. However, in our study, gaps were also identified in approximately one third of women in the vaginal and cesarean control groups, highlighting the importance of including such groups in imaging studies. Alternatively, gaps may represent normal anatomy or false-positive ultrasound findings. In fact, prior studies using imaging and histologic examination of undisturbed anatomic preparations have revealed natural gaps in the anterior external anal sphincter in up to 75% of nulliparous women.9,10 We are not aware of prior reports of internal sphincter gaps seen by ultrasonography in women who have delivered by cesarean without labor. Internal anal sphincter injury may be sustained by anal stretching such as performed for fissure-in-ano30,31 or with sexual abuse,32 but in this study, a history of prepregnancy ano-rectal surgery was an exclusion criterion. The sphincter gaps in the vaginal control group may also be explained in part by clinical misclassification. In a

recent study, reexamination and reclassification of obstetric injury by a senior clinician led to an increase in identified sphincter tears from 11% to 24.5% of 241 women after first vaginal delivery.33 We recognize several potential limitations in this study. This was a cross-sectional study of ultrasound results and symptoms occurring at 6 12 months postpartum. Because we did not collect imaging data before or immediately after the sphincter repair, we were unable to conclude whether the gaps in the sphincter tear group reflected true muscle defects, artifacts of surgery, or both. Long-term follow-up would be required to demonstrate the clinical significance of our results more distant from delivery or the impact of future pregnancies and deliveries.35,36 We defined fecal incontinence with a validated questionnaire to minimize bias and to accurately classify women with and without fecal incontinence symptoms. However, our results may underestimate the diagnostic value of endoanal ultrasonography in a population of women seeking care for fecal incontinence symptoms. In such a population, more severe fecal incontinence symptoms might be more closely associated with ultrasonographically detected sphincter gaps. Our data do not allow us to comment on the diagnostic value of ultrasonography in this setting or imply direct causation. Multiple factors influence the occurrence of a sphincter tear, and some factors may confound the association between sphincter gaps and symptoms. Therefore, we cannot conclude that the observed gaps are the direct cause of fecal incontinence symptoms in this population. Nevertheless, among women with a history of obstetric sphincter laceration, 52% of fecal incontinence was attributable to the finding of an internal sphincter gap on ultrasonography. Thus, after a first vaginal delivery associated with sphincter tear, women with a sphincter gap demonstrated with ultrasonography are more likely to have fecal incontinence symptoms than those without a gap. The remaining fecal incontinence (in women without gaps) may be due to neuromuscular impairment.37 In this research, neurophysiologic testing and functional measures of sphincter function, such as manometry, were not used to further evaluate women with an intact sphincter complex. The use of ultrasonography for the characterization of the anal sphincter complex has been shown to be reliable and sensitive, particularly for internal anal sphincter gaps when performed at a single site.8,34 The measured agreement among our experienced examiners was moderate to good ( 0.52) for internal sphincter gaps and fair ( 0.33) for external sphincter gaps. This reliability assessment was performed using static images as an exploratory review on few studies be-

VOL. 108, NO. 6, DECEMBER 2006

Richter et al

Fecal Incontinence Symptoms and Sphincter Gaps

1399

cause videotaping was not possible. We expected increased interobserver variability despite training because of the involvement of multiple investigators and review from static images, but not to this level. It has been shown that review of dynamic images of the endoanal ultrasound scan, as well as limiting the scan to the distal 1 cm of the anal canal, markedly increases reliability of diagnosis between readers.8,34 We were unable to compare dynamic sonography images, and therefore our reliability measure is probably an underestimate of the true reliability of endoanal ultrasonography. Improved reliability would likely have been found if video images were compared or if the endoanal ultrasound examination were repeated real-time by a second investigator. Strengths of this study include the multicenter design, which provided regional and racial diversity, thus increasing generalizability of our results. The participants were primiparous women with no anorectal surgery or prepregnancy fecal incontinence, so previous obstetric events or surgery could not have caused our results. Finally, and perhaps most importantly, the ultrasonographers were blinded to the womens delivery group. Few studies to date have included blinded assessments, despite the fact that lack of blinding is a wellknown cause of bias. The use of ultrasonography as a diagnostic tool in the evaluation of women with symptoms of fecal incontinence may help to determine whether a gap is present that may be amenable to surgical repair, especially in the setting of a history of obstetric sphincter tear. Conversely, if no tear is seen, patients can be directed to multidimensional behavioral therapy for symptom treatment. REFERENCES
1. Sultan AH, Kamm MA, Hudson CN, Thomas JM, Bartram CI. Anal-sphincter disruption during vaginal delivery. N Engl J Med 1993;329:190511. 2. 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;189:15439. 3. Fitzpatrick M, Behan M, OConnell PR, OHerlihy C. A randomized clinical trial comparing primary overlap with approximation repair of third-degree obstetric tears. Am J Obstet Gynecol 2000;183:12204. 4. Haadem K, Dahlstrom JA, Lingman G. Anal sphincter function after delivery: a prospective study in women with sphincter rupture and controls. Eur J Obstet Gynecol Reprod Biol 1990;35:713. 5. Haadem K, Ohrlander S, Lingman G. Long-term ailments due to anal sphincter rupture caused by delivery: a hidden problem. Eur J Obstet Gynecol Reprod Biol 1988;27:2732. 6. Borello-France D, Burgio KL, Richter HE, Zyczynski H, Fitzgerald MP, Whitehead W, et al. Fecal and urinary incontinence in primiparous women. Obstet Gynecol 2006;108:86372.

7. Pinta TM, Kylanpaa ML, Teramo KA, Luukkonen PS. Sphincter rupture and anal incontinence after first vaginal delivery. Acta Obstet Gynecol Scand 2004;83:91722. 8. Sentovich SM, Wong WD, Blatchford GJ. Accuracy and reliability of Transanal ultrasound for anterior anal sphincter injury. Dis Colon Rectum 1998;41:10004. 9. Bollard RC, Gardiner A, Lindow S, Phillips K, Duthie GS. Normal female anal sphincter: difficulties in interpretation explained. Dis Colon Rectum 2002;45:1715. 10. Fritsch H, Brenner E, Lienemann A, Ludwikowski B. Anal sphincter complex: reinterpreted morphology and its clinical relevance. Dis Colon Rectum 2002;45:18894. 11. Felt-Bersma RJ, Cuesta MA, Koorevaar M, Strijers RL, Meuwissen SG, Dercksen EJ, et al. Anal endosonography: Relationship with anal manometry and neurophysiologic tests. Dis Colon Rectum 1992;35:9449. 12. Sultan AH, Kamm MA, Talbot IC, Nicholls RJ, Bartram CI. Anal endosonography for identifying external sphincter defects confirmed histologically. Br J Surg 1994;81:4635. 13. Tiandra JJ, Milsom JW, Schroeder T, Fazio VW. Endoluminal ultrasound is preferable to electromyography in mapping anal sphincter defects. Dis Colon Rectum 1993;36:68992. 14. Sultan AH, Kamm MA, Hudson CN, Nicholls JR, Bartram CI. Endosonography of the anal sphincters: normal anatomy and comparison with manometry. Clin Radiol 1994;49:36874. 15. Sultan AH, Kamm MA, Hudson CN, Bartram CI. Thirddegree obstetric anal sphincter tears: risk factors and outcomes of primary repair. BMJ 1994;308:8879. 16. Kammerer-Doak D, Wesol AB, Rogers RG, Dominguez C, Dorin MH. A prospective cohort study of women after primary repair of obstetric anal sphincter laceration. Am J Obstet Gynecol 1999;181:131723. 17. Belmonte-Montes C, Hagerman G, Vega-Yepez PA, Hernandez-de-Anda E, Fonseca-Morales V. Anal sphincter injury after vaginal delivery in primiparous females. Dis Colon Rectum 2001;44:12448. 18. Faltin DL, Boulvain M, Irion O, Bretones S, Stan C, Weil A. Diagnosis of anal sphincter tears by postpartum endosonography to predict fecal incontinence. Obstet Gynecol 2000;95:6437. 19. Rockwood TH, Church JM, Fleshman JW, Kane RL, Mavrantonis C, Thorson AG, et al. Patient and surgeon ranking of the severity of symptoms associated with fecal incontinence. The Fecal Incontinence Severity Index. Dis Colon Rectum 1999; 42:152532. 20. Deen KI, Kumar D, Williams JG, Olliff J, Keighley MR. Anal sphincter defects: correlation between endoanal ultrasound and surgery. Ann Surg 1993;218:2015. 21. Nichols CM, Lamb EH, Ramakrishnan V. Differences in outcomes after third- versus fourth-degree perineal laceration repair: a prospective study. Am J Obstet Gynecol 2005;193:5306. 22. Sangalli MR, Floris L, Faltin D, Weil A. Anal incontinence in women with third or fourth degree perineal tears and subsequent vaginal deliveries. Aust N Z J Obstet Gynaecol 2000;40:2448. 23. Sultan AH, Kamm MA, Hudson CN, Bartram CI. Third degree obstetric anal sphincter tears: risk factors and outcome of primary repair. BMJ 1994;308:88791. 24. Pinta TM, Kylanpaa ML, Salmi TK, Teramo KA, Luukkonen PS. Primary sphincter repair: are the results of the operation good enough? Dis Colon Rectum 2004;47:1823. 25. Kammerer-Doak DN, Wesol AB, Rogers RG, Dominguez CE, Dorin MH. A prospective cohort study of women after primary repair of obstetric anal sphincter laceration. Am J Obstet Gynecol 1999;181:131723.

1400

Richter et al

Fecal Incontinence Symptoms and Sphincter Gaps

OBSTETRICS & GYNECOLOGY

26. Rieger N, Perera S, Stephens J, Coates D, Po D. Anal sphincter function and integrity after primary repair of third-degree tear: uncontrolled prospective analysis. ANZ J Surg 2004;74:1224. 27. Abramowitz L, Sobhani I, Ganansia R, Vuagnat A, Benifla JL, Darai E, et al. Are sphincter defects the cause of anal incontinence after vaginal delivery? Dis Colon Rectum 2000;43:5908. 28. Zetterstrom J, Lopez A, Holmstrom B, Nilsson BY, Tisell A, Anzen B, et al. Obstetric sphincter tears and anal incontinence: an observational follow-up study. Acta Obstet Gynecol Scand 2003;82:9218. 29. Poen AC, Felt-Bersma JF, Strijers RL, Dekker GA, Cuesta MA, Meuwissen SG. Third degree obstetric perineal tear: long-term clinical and functional results after primary repair. Br J Surg 1998;85:14338. 30. Nielson MB, Rasmussen OO, Pedersen JF, Christiansen J. Risk of sphincter damage and anal incontinence after anal dilatation for fissure-in-ano: an endoscopic study. Dis Colon Rectum 1993;36:67780. 31. Speakman CT, Burnett SJ, Kamm MA, Bartram CI. Sphincter injury after anal dilatation demonstrated by anal endosonography. Br J Surg 1991;78:142930. 32. Engel AF, Kamm MA, Bartram CI. Unwanted anal penetration as a physical cause of faecal incontinence. Eur J Gastroenterol Hepatol 1995;7:657. 33. Andrews V, Sultan AH, Thakar R, Jones PW. Occult anal sphincter injuries: myth or reality? BJOG 2006;113:195200. 34. Gold DM, Halligan S, Kmiot WA, Bartram CI. Intraobserver and interobserver agreement in anal endosonography. Br J Surg 1999;86:3715. 35. Faltin DL, Sangalli MR, Roche B, Floris L, Boulvain M, Weil A. Does a second delivery increase the risk of anal incontinence? BJOG 2001;108:6848. 36. Ryhammer AM, Bek KM, Laurberg S. Multiple vaginal deliveries increase the risk of permanent incontinence of flatus and urine in normal premenopausal women. Dis Colon Rectum 1995;38:12069. 37. Rao SS. Pathophysiology of adult fecal incontinence. Gastroenterology 2004;126:S1422.

Baylor College of Medicine Paul M. Fine, MD, Principal Investigator Rodney A. Appell, MD, Co-Principal Investigator Peter K. Thompson, MD, Co-Investigator Peter M. Lotze, MD, Co-Investigator Naomi Frierson, Research Coordinator University of Iowa Ingrid Nygaard, MD, Principal Investigator Debra Brandt, RN, Denise Haury, RN, Research Coordinators Karl Kreder, MD, Co-Investigator Catherine S. Bradley, MD, Co-Investigator Satish Rao, MD, Co-Investigator Johns Hopkins Medical Institutes Geoffrey Cundiff, MD, Principal Investigator Victoria Handa, MD, Co-Investigator Mary Elizabeth Sauter, NP, Research Coordinator Jamie Wright, MD, Co-Investigator University of Michigan Morton B. Brown, PhD, Principal Investigator John T. Wei, MD, MS, Co-Principal Investigator Beverly Marchant, RN, BS, Project Manager John O.L. DeLancey, MD, Co-Investigator Nancy K. Janz, PhD, Co-Investigator Dean G. Smith, PhD, Co-Investigator Patricia A. Wren, PhD, Co-Investigator Wei Wei, PhD, Statistician James Imus, MS, Statistician Yang Wang Casher, MS, Database Programmer University of North Carolina at Chapel Hill Anthony G. Visco, MD, Principal Investigator AnnaMarie Connolly, MD, Co-Investigator John Lavelle, MD, Co-Investigator Mary J. Loomis, RN, Research Coordinator Anita K. Murphy, NP, Research Coordinator Ellen C. Wells, MD, Co-Investigator William Whitehead, PhD, Co-Investigator Julia Fielding, MD, Co-Investigator Loyola University, Chicago Linda Brubaker, MD, Principal Investigator Mary Pat FitzGerald, MD, Co-Principal Investigator Kimberly Kenton, MD, Co-Investigator Dorothea Koch, RN, Research Coordinator Charity Ball, RN, Research Coordinator Steering Committee Chairman Robert Park, MD NICHD Project Scientist Anne M. Weber, MD, MS

APPENDIX Pelvic Floor Disorders Network Members

University of Alabama at Birmingham Holly E. Richter, PhD, MD, Principal Investigator Kathryn Burgio, PhD, Co-Principal Investigator Patricia Goode, MD, Co-Investigator R. Edward Varner, MD, Co-Investigator Gregg Shore, MD, Co-Investigator Franklin Tessler, MD, Co-Investigator Mark Lockhart, MD, MPH, Co-Investigator Velria Willis, RN, BSN, Research Coordinator University of Pittsburgh/MageeWomens Hospitals Halina Zyczynski, MD, Principal Investigator Diane Borello-France, PhD, Co-Investigator Christiane Hakim, MD, Co-Investigator Arnold Wald, MD, Co-Investigator Judy A. Gruss, BS, MS, Research Coordinator Wendy Leng, MD, Co-Investigator Pamela A. Moalli, MD, PhD, Co-Investigator

VOL. 108, NO. 6, DECEMBER 2006

Richter et al

Fecal Incontinence Symptoms and Sphincter Gaps

1401