Advertisement

Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries

Open AccessPublished:August 21, 2020DOI:https://doi.org/10.1016/j.ajog.2020.08.051

      Background

      The long-term effects of 1 or 2 consecutive obstetrical anal sphincter injuries on bowel continence are still inadequately investigated, and published results remain contradictory.

      Objective

      This study aimed to present detailed descriptive measures of the current bowel incontinence 20 years after the first birth in women who had 2 vaginal deliveries with and without sphincter injuries.

      Study Design

      Birth register data were used prospectively and linked to information from a questionnaire survey about current symptoms. Women with 2 singleton vaginal births, from 1992 to 1998, and no further births were retrieved and surveyed by the Swedish Medical Birth Register and Statistics Sweden in 2015. A simple random sample of 11,000 women was drawn from a source cohort of 64,687 women. The cumulative effect was studied in all women with a repeat sphincter injury from 1987 to 2000. Postal and web-based questionnaires were used. The study population consisted of 6760 women with no sphincter injury, 357 with 1 sphincter injury, and 324 women with 2 sphincter injuries. Women with 2 deliveries without sphincter injuries aged 40 to 60 years as reference, were compared with those of women that sustained 1 or 2 consecutive sphincter injuries. Here, third- and fourth-degree perineal tears were presented as 1 group. Fecal incontinence was defined as current involuntary leakage of solid or liquid stool, with and without concomitant leakage of gas. The Fisher exact test and the Mann-Whitney U test were used to compare the results of the 2 groups. The trend was analyzed using the Mantel-Haenszel statistics. Logistic regression models obtained the estimated age-related probability of fecal incontinence components.

      Results

      The risk of sphincter injury at first delivery was 3.9%, and the risk of a repeat sphincter injury was 10.0% (odds ratio, 2.70; 95% confidence interval, 1.80–4.07). The overall prevalence of fecal incontinence in women without sphincter injuries was 11.7%, which doubled to 23.8% (odds ratio, 2.27; 95% confidence interval, 1.75–2.94) in those with 1 sphincter injury and more than tripled to 36.1% (odds ratio, 3.97; 95% confidence interval, 3.11–5.07) after 2 sphincter injuries (trend P<.0001). The proportion of women with severe fecal incontinence increased 3-fold and 5-fold from 1.8% after no obstetrical anal sphincter injury to 5.4% (95% confidence interval, 3.3–8.2) and 9.0% (95% confidence interval, 6.1–12.6) after 1 or 2 obstetrical anal sphincter injuries, respectively (trend P<.0001). In women without sphincter injuries, the estimated probability of fecal incontinence increased from 7.0% at the age of 40 years to 19.8% at the age of 60 years. In contrast, in women with 1 or 2 sphincter injuries, the estimated probability of fecal incontinence increased from 26.1% and 33.3%, respectively, at the age of 40 years to 36.8% and 48.8% at the age of 60 years. The prevalence of fecal incontinence increased after 52 years of age in women with 1 or 2 sphincter injuries. The dominant types of leakage in women with fecal incontinence were the combination of liquid stool and gas, and the triple combination consisting of solid and liquid stools and gas. The triple combination increased from 18.9% in those without sphincter injury to 28.2% in women with 2 injuries (trend P=.0204).

      Conclusion

      The risk of sustaining a repeat sphincter injury at the second delivery was almost tripled compared with the risk at the first delivery. Furthermore, 1 or 2 sphincter injuries brought severe long-term consequences for bowel continence. Accidental leakage of stool and gas increased with each sphincter injury, and the effect was proportionally cumulative. After the age of 52 years, the prevalence of fecal incontinence seemed to accelerate.

      Key words

      Introduction

      The long-term consequences of childbirth should attract growing interest in epidemiology as women nowadays live longer, healthier, and more active lives. An obstetrical anal sphincter injury (OASI) is an appalling trauma to the pelvic floor sustained during the second stage of labor. This perineal laceration, even if adequately sutured, poses a substantial threat to bowel continence.
      ,
      • Bols E.M.
      • Hendriks E.J.
      • Berghmans B.C.
      • Baeten C.G.
      • Nijhuis J.G.
      • De Bie R.A.
      A systematic review of etiological factors for postpartum fecal incontinence.
      Furthermore, a primary repair may fail because of infection and wound breakdown in up to 30% of cases.
      • Kirss J.
      • Pinta T.
      • Böckelman C.
      • Victorzon M.
      Factors predicting a failed primary repair of obstetric anal sphincter injury.
      Most OASIs occur seemingly by chance in the absence of known risk markers, and there is still no prediction model that is of use to avoid OASI in the clinical setting.
      • Webb S.S.
      • Hemming K.
      • Khalfaoui M.Y.
      • et al.
      An obstetric sphincter injury risk identification system (OSIRIS): is this a clinically useful tool?.
      Therefore, these injuries are often excused as inevitable and impossible to foresee. According to the Swedish Medical Birth Register (MBR), OASIs occurred in 6.6%, 2.3%, and 0.9% in the first, second, and third births from 1999 to 2011, indicating that vaginal parity is a strong predictor.
      • Waldenström U.
      • Ekéus C.
      Risk of obstetric anal sphincter injury increases with maternal age irrespective of parity: a population-based register study.
      The reported national rate of OASI, according to the MBR in primiparous women, increased considerably from 2.9% in 1990 to 7.0% in 2004 (Supplemental Figure 1; Supplemental Tables 1 and 2).
      The Swedish National Board of Health and Welfare
      Statistics database for pregnancies, births and newborns. Socialstyrelsen.
      A similar trend was noticed in other European countries during the 1990s.
      • Ekéus C.
      • Nilsson E.
      • Gottvall K.
      Increasing incidence of anal sphincter tears among primiparas in Sweden: a population-based register study.
      • Laine K.
      • Gissler M.
      • Pirhonen J.
      Changing incidence of anal sphincter tears in four Nordic countries through the last decades.
      • Gurol-Urganci I.
      • Cromwell D.A.
      • Edozien L.C.
      • et al.
      Third- and fourth-degree perineal tears among primiparous women in England between 2000 and 2012: time trends and risk factors.
      The introduction of endoanal ultrasonography in 1989 made it possible to detect signs of OASIs, unnoticed or unreported at birth, in 1 of 3 women.
      • Sultan A.H.
      • Kamm M.A.
      • Hudson C.N.
      • Thomas J.M.
      • Bartram C.I.
      Anal-sphincter disruption during vaginal delivery.
      This insight led to the introduction of a standardized classification of perineal tears,
      • Sultan A.H.
      Editorial: obstetric perineal injury and anal incontinence.
      the implementation of new policies, the training of staff in recognizing OASIs, and an increased awareness of a previously neglected complication.
      • Laine K.
      • Gissler M.
      • Pirhonen J.
      Changing incidence of anal sphincter tears in four Nordic countries through the last decades.
      A possible link between OASI and bowel incontinence is still disputed. In a recent review by LaCross et al,
      • LaCross A.
      • Groff M.
      • Smaldone A.
      Obstetric anal sphincter injury and anal incontinence following vaginal birth: a systematic review and meta-analysis.
      it was stated that almost every second report negated an association. A broad spectrum of demoralizing consequences of fecal incontinence (FI) after OASI has been described in detail in qualitative studies,
      • Olsson F.
      • Berterö C.
      Living with faecal incontinence: trying to control the daily life that is out of control.
      revealing various socially incapacitating disabilities and severe constraints on daily life activities, often unvoiced in isolation.
      • Rasmussen J.L.
      • Ringsberg K.C.
      Being involved in an everlasting fight--a life with postnatal faecal incontinence. A qualitative study.
      ,
      • Johanson J.F.
      • Lafferty J.
      Epidemiology of fecal incontinence: the silent affliction.
      Despite its severe implications, 70% to 90% of women with FI did not seek medical advice and treatment.
      • Meyer I.
      • Richter H.E.
      Impact of fecal incontinence and its treatment on quality of life in women.
      There are still considerable gaps in knowledge about the long-term effects of OASI and the possible cumulative effect of a second OASI on FI. A complicating circumstance is that the median age for the onset of FI has been reported to be 55 years.
      • Bharucha A.E.
      • Zinsmeister A.R.
      • Locke G.R.
      • et al.
      Prevalence and burden of fecal incontinence: a population-based study in women.
      This study aimed to determine the risk of a recurrent OASI and the long-term age-related prevalence of FI in women aged 40 to 60 years after 1 or 2 OASIs compared with those in women without OASI.

       Why was this study conducted?

      The long-term effects on bowel continence of 1 or 2 obstetrical sphincter injuries are still inadequately investigated, and published results remain contradictory.

       Key findings

      After a sphincter injury at the first birth, the risk of a repeat injury almost tripled. The overall prevalence of fecal incontinence (FI) in women without sphincter injuries was 11.7%, which doubled to 23.8% in those with 1 sphincter injury and more than tripled to 36.1% after 2 sphincter injuries.

       What does this add to what is known?

      The effect of a repeat sphincter injury was proportionally cumulative. After 1 or 2 injuries, the incidence of FI increased after the age of 52 years. After sphincter injuries, 1 in 5 remained completely continent at 60 years of age.

      Materials and Methods

      Ethical approval for this study was obtained from the Regional Ethical Review Board (reference number, 776-3; November 18, 2013, Gothenburg, Sweden). All women gave their written consent to participate.

       The study population

      This report is 1 part of the national Swedish Pregnancy, Obesity, and Pelvic floor survey on women who had 2 vaginal deliveries. Birth register data were used prospectively and combined with information from a survey from January 2015 to May 2015 about current symptoms of pelvic floor disorders (PFDs), 2 decades after childbirth. Inclusion criteria were 2 consecutive singleton vaginal deliveries (VD) from 1992 to 1998 with no previous or further births and no ongoing pregnancy. Eligible women (n=64,687) were identified and retrieved by the MBR, (https://www.socialstyrelsen.se/en/statistics-and-data/registers/register-information/the-swedish-medical-birth-register/) and the Total Population Register (TPR) by Statistics Sweden (https://www.scb.se/en/finding-statistics/statistics-by-subject-area/population/population-composition/population-statistics/) (Figure 1, A). From this cohort, a simple random sample of 11,000 women was drawn and surveyed by Statistics Sweden. Postal and web-based questionnaires were used. After 3 mailing cycles, the forms were returned by 7177 women (65.5%). Of the 7177 women, 32 were excluded for various reasons. In addition, among 7145 women with 2 VDs from the random sample, 6760 had no OASI, 357 had 1 OASI (253 at the first delivery and 104 at the second delivery), and 28 had 2 OASIs (Figure 1, A). To compare the effect of 1 OASI vs 2 OASIs, a cohort of all women with 2 consecutive OASIs from 1987 to 2000 was retrieved by the MBR. The inclusion criteria were otherwise identical to those in the randomly sampled group. The number of eligible women with 2 OASIs was 469, of which 39 were oversubscribed (Figure 1, B). After the exclusion of nonresponders and other losses, the total study cohort was 7441 women, 7117 women from the randomly selected cohort from 1992 to 1998 and 324 women with 2 OASIs from 1987 to 2000 (Figure 1). Given the size of the study cohorts, an alpha level of 0.05, and a power value of 80% and using the Fisher exact test for the analysis, power calculations were performed to determine the minimum significant difference in the prevalence of FI between cohorts (Supplemental Figure 2). Details about demographics of responders vs nonresponders are presented in Supplemental Table 3.
      Figure thumbnail gr1
      Figure 1Flowchart of the 2 cohorts of women in the study
      A, Cohort 1, a random sample of 11,000 women from 1992 to 1998. B, Cohort 2, all women with repeat (2) OASIs from 1987 to 2000. aThe source cohort denotes all women who had 2 vaginal deliveries registered in Sweden with 2 singleton births from the period 1992 to 1998. bMisclassification for parity, deceased, emigrated, and change of social security number was made by Statistics Sweden, which performed an ID run in January 12, 2014, to exclude these categories. This information is updated every fourth week. cMail hindrance denotes mail returned to sender and addressee and those not found or emigrated. dExcluded denotes declined participation and a blank or an unusable form. eAmong women with 1 OASI, 253 had an OASI at the first birth and 104 had an OASI at the second birth. fThese 28 women occurred both in the random sample and in the 2-OASI cohort. gAll women who had 2 vaginal deliveries and had sustained 2 consecutive OASIs from the period 1987 to 2000.
      ID, identification; OASI, obstetrical anal sphincter injury.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.

       Definition of sphincter injury

      Third- and fourth-degree injuries, according to the classification of Sultan,
      • Sultan A.H.
      Editorial: obstetric perineal injury and anal incontinence.
      were presented as 1 group and identified by codes 664.2 and 664.3 in the International Classification of Diseases, ninth revision (ICD-9) (1987–1996), and O70.2 and O70.3 in the International Classification of Diseases, 10th revision. From the period 1997 to 2000, OASI was also identified using the surgical code MBC33.

       The questionnaire and definition of outcomes

      The questionnaire consisted of 40 questions and is available in the Appendix. The introductory section was about demographics—current height and weight, menstruation, ongoing pregnancy, hysterectomy, etcetera—and a control question about parity. The second section covered validated questions about PFDs.
      • Sandvik H.
      • Hunskaar S.
      • Seim A.
      • Hermstad R.
      • Vanvik A.
      • Bratt H.
      Validation of a severity index in female urinary incontinence and its implementation in an epidemiological survey.
      • Tegerstedt G.
      • Miedel A.
      • Maehle-Schmidt M.
      • Nyren O.
      • Hammarström M.
      A short-form questionnaire identified genital organ prolapse.
      • Jorge J.M.
      • Wexner S.D.
      Etiology and management of fecal incontinence.
      FI was defined according to the International Urogynecological Association (IUGA) and the International Continence Society (ICS) as involuntary leakage of solid and liquid stool, with and without concomitant leakage of gas.
      • Haylen B.T.
      • De Ridder D.
      • Freeman R.M.
      • et al.
      An International Urogynecological Association (IUGA)/International Continence Society (ICS) joint report on the terminology for female pelvic floor dysfunction.
      Anal incontinence (AI) was defined as FI or isolated gas leakage. Severe FI or AI was defined as a condition that was bothersome FI or AI (some bother, much bother, a major problem, but not a minor nuisance) occurring several times a month or more often.
      Any solid or liquid leakage refers to the symptom isolated or any combination in women having FI. Body mass index (BMI) was calculated using the current weight and height.

       Statistics

      Continuous variables were presented as mean and standard deviation and as median and interquartile range. Categorical data were presented as number, percent, and 95% confidence interval (CI). When comparing cohort characteristics between 2 groups, the Fisher exact test was used for dichotomous variables and the Mann-Whitney U test for continuous variables. The results were presented as the mean difference for continuous variables and as the difference in percentages for categorical variables, 95% CI, and P value. Calculations of the 95% CI for the difference among continuous variables were based on bootstrapping of 10,000 replicates, identifying the 2.5 and 97.5 percentiles of the mean difference. The calculation of the 95% CI for the difference in percentages among the categorical variables was based on the exact method. The prevalence was calculated for all women in the study cohorts. When comparing the prevalence between 2 groups, logistic regression models were used, adjusting for age and current BMI. Results were presented as adjusted odds ratio (aOR), 95% CI, and P value. The trend for dichotomous variables among the cohorts with no OASI, 1 OASI, and 2 OASIs was analyzed using the Mantel-Haenszel and Pearson chi-squared statistics. For continuous variables, the trend was analyzed using the Spearman rank correlation test. For the analysis of the trend, the order among the cohorts assumed that women without OASI were least affected, followed by those with 1 OASI, and that women with 2 OASIs were most adversely affected. The estimated age-related probability for FI was obtained from 3 (1 for each OASI group) independent logistic regression models. Because the observed frequency of OASI increased from 3% in 1987 to 6.7% in 2000 (assuming the 1987 rate as underestimated), the logistic models were weighted with the year 2000 as reference. Results were presented as aORs per 10 years, 95% CI, and P value. The estimated probability of FI by age was presented along with the 95% CI. In addition, for the 1- and 2-OASI cohorts, segmented logistic regressions were performed with 52 years of age as a cutoff. This cutoff was based on the observed age-related values of FI. In each analysis, missing data were excluded. No adjustment was made for multiple testing. All statistical testing was 2-sided, and the significance level was set to a P value of <.05. Statistical analyses were performed using Statistical Analysis System (SAS version 9.4; SAS Institute, Cary, NC).

      Results

       Cohort characteristics

      The response rate was 65.5% in the randomly selected group and 70.1% among those with 2 OASIs. The web-based questionnaire was used by 56.2%. The overall rate of missing data for outcomes was low, ranging from 0.3% (maternal height) to 2.2% (causes for amenorrhea). The risk of OASI at the first delivery was 3.9% (281/7145), and the risk of a repeat OASI was 10.0% (28/281) (OR, 2.70; 95% CI, 1.80–4.07) (Not shown in Table). Infant birthweight ≥ 4 kg was almost doubled in the OASI cohorts in both the first and the second deliveries (trend P<.0001) (Table 1). The vacuum extraction rate was more than doubled in the OASI cohorts and was used 5 to 6 times more often in the first delivery than in the second delivery (Table 1). Women with no OASI were somewhat younger and, therefore, less often postmenopausal. The follow-up time from first birth was approximately 0.6 years longer in women with 2 OASIs, reflecting the extended inclusion period, but the time between births was almost the same (Table 1). An OASI at the first delivery prolonged the time to the next by approximately 0.2 years (P=.0006) (Supplemental Figure 3; Supplemental Table 4).
      Table 1Cohort characteristics
      VariableStudy cohortsTrend P value
      Trend was analyzed using the Spearman rank correlation test and Mantel-Haenszel chi-squared statistics
      Difference among cohorts
      No OASI (n=6760)1 OASI (n=357)2 OASIs (n=324)No vs 1No vs 21 vs 2
      Mean (SD)

      Median (IQR)
      Mean (95% CI);

      P value
      Age, y49.1 (4.0)

      48.9 (46.4–51.5)
      49.8 (4.1)

      49.5 (47.1–52.2)
      50.6 (4.2)

      50.2 (47.5–53.3)
      <.0001−0.71 (−1.14 to −0.28)

      .0017
      −1.52 (−2.00 to −1.06)

      <.0001
      −0.82 (−1.45 to −0.20)

      .027
      Current weight, kg70.5 (12.3)

      69.0 (62.0–77.0)
      71.2 (13.3)

      69.0 (62.0–77.5)
      70.5 (12.3)

      69.0 (62.0–78.0)
      .61−0.73 (−2.18 to 0.68)

      .57
      −0.01 (−1.40 to 1.36)

      .86
      0.72 (−1.20 to 2.65)

      .77
      Height, cm167.4 (5.9)

      168.0 (163.0–171.0)
      166.8 (6.0)

      167.0 (163.0–171.0)
      167.1 (5.8)

      167.0 (163.0–171.0)
      .0760.58 (−0.06 to 1.22)

      .13
      0.26 (−0.40 to 0.90)

      .28
      −0.32 (−1.17 to 0.57)

      .79
      Current BMI, kg/m225.1 (4.2)

      24.3 (22.2–27.3)
      25.6 (4.7)

      24.4 (22.5–27.6)
      25.2 (4.0)

      24.4 (22.3–27.5)
      .20−0.45 (−0.96 to 0.04)

      .24
      −0.04 (−0.49 to 0.41)

      .49
      0.41 (−0.24 to 1.07)

      .77
      Time from first delivery, y21.6 (1.3)

      21.7 (20.6–22.7)
      21.5 (1.3)

      21.7 (20.6–22.5)
      22.1 (2.9)

      22.1 (19.8–24.2)
      .0570.03 (−0.11 to 0.170)

      .66
      −0.57 (−0.89 to −0.24)

      .0019
      −0.60 (−0.94 to −0.25)

      .013
      Time between first and second delivery, y2.6 (1.0)

      2.4 (1.9–3.2)
      2.7 (1.0)

      2.5 (2.0–3.2)
      2.9 (1.3)

      2.6 (2.0–3.4)
      .0006−0.10 (−0.21 to 0.01)

      .097
      −0.28 (−0.42 to −0.14)

      .0011
      −0.18 (−0.36 to −0.004)

      .22
      N (%)

      95% CI
      Trend P valuePercentage (95% CI)

      P value
      Birthweight >4 kg at first delivery904 (13.4)

      12.6–14.2
      84 (23.5)

      19.2–28.3
      88 (27.2)

      22.4–32.4
      <.000110.2 (5.5–14.8)

      <.0001
      13.8 (8.7–18.9)

      <.0001
      3.6 (−3.2 to 10.5)

      .29
      Birthweight >4 kg at second delivery1594 (23.6)

      22.6–24.6
      131 (36.7)

      31.7–41.9
      154 (47.5)

      42.0–53.1
      <.000113.1 (7.9–18.4)

      <.0001
      24.0 (18.3–29.6)

      <.0001
      10.8 (3.2–18.5)

      .0051
      Vacuum extraction at first delivery845 (12.5)

      11.7–13.3
      92 (25.8)

      21.3–30.6
      104 (32.1)

      27.1–37.5
      < .000113.3 (8.5–18.0)

      <.0001
      19.6 (14.3–24.9)

      <.0001
      6.3 (−0.8 to 13.4)

      .075
      Vacuum extraction at second delivery143 (2.1)

      1.8–2.5
      18 (5.0)

      3.0–7.9
      21 (6.5)

      4.1–9.7
      <.00012.9 (0.5; 5.4)

      .0013
      4.4 (1.5–7.2)

      <.0001
      1.4 (−2.4 to 5.2)

      .51
      Current BMI ≥30, kg/m2849 (12.6)

      11.9–13.5
      58 (16.5)

      12.8–20.8
      34 (10.6)

      7.5–14.5
      .983.8 (−0.3 to 7.9) 0.041−2.0 (−5.6 to 1.6)

      0.34
      −5.9 (−11.3 to −0.4)

      .033
      Hysterectomy263 (3.9)

      3.4–4.4
      12 (3.4)

      1.8–5.8
      16 (4.9)

      2.9–7.9
      .53−0.5 (−2.6 to 1.5)

      .78
      1.0 (−1.5 to 3.6)

      .31
      1.6 (−1.7 to 4.9)

      .34
      Postmenopausal1533 (22.7)

      21.7–23.7
      105 (29.4)

      24.7–34.4
      98 (30.2)

      25.3–35.6
      <.00016.7 (1.8–11.7) .00447.6 (2.3–12.8)

      .0023
      0.8 (−6.3 to 8.0)

      .87
      Estrogen treatment
      Both systemic and vaginal treatments
      144 (2.1)

      1.8–2.5
      2 (0.6)

      0.1–2.0
      5 (1.5)

      0.5–3.6
      .13−1.6 (−2.6 to −0.6)

      .034
      −0.6 (−2.1 to 1.0)

      .69
      1.0 (−0.9 to 2.8)

      .27
      Childhood nocturnal enuresis
      ≥5 years of age.
      506 (7.5)

      6.9–8.2
      27 (7.6)

      5.1–10.9
      28 (8.7)

      5.9–12.3
      .480.1 (−2.9 to 3.1)

      .92
      1.2 (−2.1 to 4.5)

      .45
      1.1 (−3.4 to 5.5)

      .67
      BMI, body mass index; CI, confidence interval; IQR, interquartile range; OASI, obstetrical anal sphincter injury; SD, standard deviation.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      a Trend was analyzed using the Spearman rank correlation test and Mantel-Haenszel chi-squared statistics
      b Both systemic and vaginal treatments
      c ≥5 years of age.

       Incontinence components

      The overall prevalence of FI in women without OASI was 11.7%, which doubled to 23.7% (OR, 2.27; 95% CI, 1.75–2.94) in those with 1 OASI and more than tripled to 36.1% (OR, 3.97; 95% CI, 3.11–5.07) after 2 OASIs (trend P<.0001) (Table 2). Severe FI increased 3-fold and 5-fold from 1.8% after no OASI to 5.4% (95% CI, 3.3–8.2) and 9.0% (95% CI, 6.1–12.6) after 1 or 2 OASIs, respectively (trends P<.0001) (Table 2). The same pattern was observed for severe AI. Women with severe FI accounted for 15% of those with FI and no OASI and 25% of those with 2 OASIs (not indicated in tables). Except for isolated gas incontinence, this trend was present for all components of bowel incontinence (Figure 2). The effect size on the difference among the cohorts was smaller for AI than those for FI (Table 2). At 60 years of age, less than 20% of the women with OASI were completely continent compared with 32% of the women without OASI (Supplemental Tables 5 and 6). The dominant types of leakage in women with FI were the combination of liquid stool and gas and the triple combination consisting of solid and liquid stools and gas. The triple combination among women with FI increased from 18.9 in women without OASI to 28.2% after 2 OASIs (trend P=.0204) (Table 3; Figure 3). The prevalence of the triple combination increased from 2.2% (95% CI, 1.9–2.6) to 10.2% (95% CI, 7.4–14.0) in those with 2 OASIs (not shown in tables). The 3 main components of bowel leakage, single and in combinations, that is, “any solid stool,” “any liquid stool,” and “any gas,” increased with a higher number of OASI. “Any solid stool” increased from 3.4% (no OASI) to 11.7% (2 OASIs) (trend P<.0001). “Any liquid stool” and “any gas” increased from 10.8% and 9.8% to 34.9% and 34.3%, respectively, in women without OASI compared with those with 2 OASIs (trend P<.0001) (Table 2; Figure 2). The trajectories of the observed prevalence indicated a cutoff at 52 years of age after which the incidence of FI increased (Figure 4).
      Table 2Components of bowel incontinence
      VariableStudy cohortsTrend
      Trend was analyzed with Mantel-Haenszel statistics
      P value
      Difference among cohorts
      No OASI (n=6717)

      n (%)

      95% CI
      1 OASI (n=355)

      n (%)

      95% CI
      2 OASIs (n=324)

      n (%)

      95% CI
      No vs 1

      aOR
      aOR comparison among independent groups was calculated from a logistic regression model with adjustment for current age and BMI
      (95% CI)

      P value
      No vs 2

      aOR (95% CI)

      P value
      1 vs 2

      aOR (95% CI)

      P value
      FI784 (11.7)

      10.9–12.5
      84 (23.7)

      19.3–28.4
      117 (36.1)

      30.9–41.6
      <.00012.27 (1.75–2.94)

      <.0001
      3.97 (3.11–5.07)

      <.0001
      1.78 (1.27–2.51)

      .0009
      Any solid stool
      Any solid or liquid leakage refers to the symptom isolated or any combination in women having FI
      229 (3.4)

      3.0–3.9
      26 (7.3)

      4.8–10.6
      38 (11.7)

      8.4–15.7
      <.00012.19 (1.43–3.33)

      .0003
      3.59 (2.49–5.18)

      <.0001
      1.68 (0.99–2.84)

      .055
      Any liquid stool724 (10.8)

      10.1–11.5
      77 (21.7)

      17.5–26.4
      113 (34.9)

      29.7–40.3
      <.00012.21 (1.69–2.89)

      <.0001
      4.14 (3.23–5.30)

      <.0001
      1.90 (1.34–2.69)

      .0003
      Concomitant gas
      Concomitant gas denotes leakage of gas in combination with solid and liquid stools, that is, in women with FI
      656 (9.8)

      9.1–10.5
      78 (22.0)

      17.8–26.6
      111 (34.3)

      29.1–39.7
      <.00012.52 (1.93–3.29)

      <.0001
      4.46 (3.48–5.71)

      <.0001
      1.77 (1.25–2.51)

      .0012
      AI3509 (52.2)

      51.0–53.4
      245 (69.0)

      63.9–73.8
      249 (76.9)

      71.9–81.3
      <.00011.99 (1.57–2.51)

      <.0001
      2.77 (2.13–3.61)

      <.0001
      1.41 (1.00–1.99)

      .052
      Completely continent
      Completely continent means not having either FI or isolated gas incontinence
      3208 (47.8)

      46.6–49.0
      110 (31.0)

      26.2–36.1
      75 (23.1)

      18.7–28.1
      <.0001
      Isolated gas incontinence
      Isolated gas incontinence does not include leakage of gas in combination with solid or liquid stool, that is, as a component of FI
      2725 (40.6)

      39.4–41.8
      161 (45.4)

      39.4–49.7
      132 (40.7)

      35.5–46.2
      .411.19 (0.96–1.48)

      .11
      0.96 (0.76–1.21)

      .71
      0.81 (0.60–1.11)

      .20
      Severe FI
      Severe FI or AI was defined as having bothersome FI or AI several times a month or more often.
      122 (1.8)

      1.5–2.2
      19 (5.4)

      3.3–8.2
      29 (9.0)

      6.1–12.6
      <.00012.84 (1.72–4.68)

      <.0001
      4.56 (2.95–7.05)

      <.0001
      1.60 (0.87–2.94)

      .13
      Severe AI336 (5.0)

      4.5–5.6
      39 (11.0)

      7.9–14.7
      66 (20.4)

      16.1–5.2
      <.00012.24 (1.58–3.20)

      <.0001
      4.37 (3.25–5.89)

      <.0001
      1.91 (1.24–2.95)

      .0036
      AI, anal incontinence; aOR, adjusted odds ratio; CI, confidence interval; FI, fecal incontinence; OASI, obstetrical anal sphincter injury.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      a Trend was analyzed with Mantel-Haenszel statistics
      b aOR comparison among independent groups was calculated from a logistic regression model with adjustment for current age and BMI
      c Any solid or liquid leakage refers to the symptom isolated or any combination in women having FI
      d Concomitant gas denotes leakage of gas in combination with solid and liquid stools, that is, in women with FI
      e Completely continent means not having either FI or isolated gas incontinence
      f Isolated gas incontinence does not include leakage of gas in combination with solid or liquid stool, that is, as a component of FI
      g Severe FI or AI was defined as having bothersome FI or AI several times a month or more often.
      Figure thumbnail gr2
      Figure 2Components of anal and fecal incontinence
      Differences among independent groups were analyzed using the Mantel-Haenszel statistics. The calculation of the 95% CI for the difference in percentages among categorical variables was based on the exact method.
      CI, confidence interval; OASI, obstetrical anal sphincter injury.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      Table 3Different types of incontinence
      Types: isolated and in combinationsNo OASI1 OASI2 OASIsTrend
      Trend was analyzed with Mantel-Haenszel statistics, whereas the distribution among cohorts according to all types of presentation (a–f) was analyzed with the Pearson chi-squared test
      P value
      N=6717N=355N=324
      n (%)95% CIn (%)95% CIn (%)95% CI
      • a.
        Solid stool only
      16 (2.0)1.2–3.31 (1.2)0.0–6.50 (0.0).0409
      Value analyzed using the Pearson chi-squared test
      • b.
        Liquid stool only
      91(11.6)9.5–14.14 (4.8)1.3–11.85 (4.3)1.4–9.7
      • c.
        Solid+liquid
      21 (2.7)1.7–4.11 (1.2)0.0–6.51 (0.9)0.0–4.7
      • d.
        Solid+gas
      44 (5.6)4.1–7.56 (7.1)2.7–14.94 (3.4)0.9–8.5
      • e.
        Liquid+gas
      464 (59.2)55.7–62.754 (64.3)53.1–74.574 (63.3)53.8–72.0
      • f.
        Solid+liquid+gas
        The triple combination, leakage for solid and liquid stools and gases, increased significantly in women with FI and with no OASI, 1 OASI, and 2 OASIs. Conversely, the combined rate of the 4 rare types of leakage decreased correspondingly with a trend that was significant (P = .0003)
      148 (18.9)16.2–21.818 (21.4)13.2–31.733 (28.2)20.3–37.3
      FI (a–f)
      a–f indicates 6 possible single or combined presentations obtained from the answers in the questionnaire, which sum up to the generic term “fecal incontinence”
      784 (100.0)84 (100.0)117 (100.0)
      Any solid stool (a+c+d+f)
      Any solid or liquid leakage refers to the symptom isolated or in any combination in women having FI
      229 (3.4)3.0–3.926 (7.3)4.8–10.638 (11.7)8.4–15.7<.0001
      Any liquid stool (b+c+e+f)724 (10.8)10.1–11.577 (21.7)17.5–26.4113 (34.9)29.7–40.3<.0001
      Concomitant gas
      Concomitant gas indicates leakage of gas in combination with solid and liquid stools, that is, in women with FI (Figure 3).
      (d+e+f)
      656 (9.7)9.1–10.578 (22.0)17.8–26.6111 (34.3)29.1–39.7<.0001
      The 2 dominant combinations (e+f)612 (78.1)75.0–80.972 (85.7)76.4–92.4107 (91.5)84.8–95.8.0003
      The 4 rare single and combined presentations (a+b+c+d)172 (21.9)19.1–25.012 (14.3)7.6–23.610 (8.5)4.2–15.2
      Missing values in the analysis of prevalence of components of FI are as follows: 43 (no OASI), 2 (1 OASI), and 0 (2 OASIs).
      CI, confidence interval; FI, fecal incontinence; OASI, obstetrical anal sphincter injury.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      a Trend was analyzed with Mantel-Haenszel statistics, whereas the distribution among cohorts according to all types of presentation (a–f) was analyzed with the Pearson chi-squared test
      b Value analyzed using the Pearson chi-squared test
      c The triple combination, leakage for solid and liquid stools and gases, increased significantly in women with FI and with no OASI, 1 OASI, and 2 OASIs. Conversely, the combined rate of the 4 rare types of leakage decreased correspondingly with a trend that was significant (P = .0003)
      d a–f indicates 6 possible single or combined presentations obtained from the answers in the questionnaire, which sum up to the generic term “fecal incontinence”
      e Any solid or liquid leakage refers to the symptom isolated or in any combination in women having FI
      f Concomitant gas indicates leakage of gas in combination with solid and liquid stools, that is, in women with FI (Figure 3).
      Figure thumbnail gr3
      Figure 3Single and combinations of components of fecal incontinence
      The trend was analyzed using the Mantel-Haenszel statistics and was significant (P=.0204) for the increase of the triple combination and for the decrease of the sum of the rarely occurring symptoms. aTriple combination of symptoms; bSum of the rarely occurring symptoms.
      OASI, obstetrical anal sphincter injury.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      Figure thumbnail gr4
      Figure 4The age-related change of fecal incontinence
      A, Panel A indicates the observed values of FI stratified according to age groups. The error bars denote the 95% CI. B, Panel B indicates the age-related estimated probability of FI, from age 40 to 60 years, from 3 independent logistic regression models. C, Panel C indicates a segmented logistic regression with a cutoff at 52 years of age for women with 1 OASI and women with 2 OASIs as the incidence of FI seemed to accelerate after this point. The shaded areas show the 95% 2-tailed CIs for the estimated probability.
      CI, confidence interval; FI, fecal incontinence; OASI, obstetrical anal sphincter injury; OR, odds ratio.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.

      Comments

       Principle findings

      After a sphincter injury at the first birth, the risk of having a repeat injury almost tripled. Compared with women who had 2 vaginal deliveries without OASI, the prevalence of all components of FI was doubled or tripled 2 decades after 1 or 2 OASIs. Severe FI increased 3-fold and 5-fold after 1 or 2 OASIs, respectively. In women without OASI, the age-related incidence of FI was rather constant across the ages of 40 to 60 years, whereas in women with sphincter injury, the incidence increased after the age of 52 years. The triple combination, solid and liquid stools and gas, increased after 1 or 2 sphincter injuries. At the age of 60 years, less than 1 in 5 women with OASI was continent.

       Results in context

      In a review of Jha and Parker
      • Jha S.
      • Parker V.
      Risk factors for recurrent obstetric anal sphincter injury (rOASI): a systematic review and meta-analysis.
      on 99,042 women, the mean risk of a repeat OASI was 6.3% with a range of 2.0% to 13.4%. This wide range was thought to be because of the skewed distribution of risk factors of OASI and obstetrical practices. Our result (10.0%) may be somewhat inflated because of the higher response rate after 2 OASIs (70.1% vs 65.5%).
      At present, most studies about long-term FI after OASI have used convenience samples from single hospitals
      • Faltin D.L.
      • Otero M.
      • Petignat P.
      • et al.
      Women’s health 18 years after rupture of the anal sphincter during childbirth: I. Fecal incontinence.
      • Baud D.
      • Meyer S.
      • Vial Y.
      • Hohlfeld P.
      • Achtari C.
      Pelvic floor dysfunction 6 years post-anal sphincter tear at the time of vaginal delivery.
      • Wegnelius G.
      • Hammarström M.
      Complete rupture of anal sphincter in primiparas: long-term effects and subsequent delivery.
      • Evers E.C.
      • Blomquist J.L.
      • McDermott K.C.
      • Handa V.L.
      Obstetrical anal sphincter laceration and anal incontinence 5-10 years after childbirth.
      • Soerensen M.M.
      • Buntzen S.
      • Bek K.M.
      • Laurberg S.
      Complete obstetric anal sphincter tear and risk of long-term fecal incontinence: a cohort study.
      • Halle T.K.
      • Salvesen K.Å.
      • Volløyhaug I.
      Obstetric anal sphincter injury and incontinence 15-23 years after vaginal delivery.
      and a follow-up time of <10 years.
      • Baud D.
      • Meyer S.
      • Vial Y.
      • Hohlfeld P.
      • Achtari C.
      Pelvic floor dysfunction 6 years post-anal sphincter tear at the time of vaginal delivery.
      • Wegnelius G.
      • Hammarström M.
      Complete rupture of anal sphincter in primiparas: long-term effects and subsequent delivery.
      • Evers E.C.
      • Blomquist J.L.
      • McDermott K.C.
      • Handa V.L.
      Obstetrical anal sphincter laceration and anal incontinence 5-10 years after childbirth.
      None had full control over the mode or number of previous and subsequent births. In addition, 2 studies included only women with complete OASI,
      • Wegnelius G.
      • Hammarström M.
      Complete rupture of anal sphincter in primiparas: long-term effects and subsequent delivery.
      ,
      • Soerensen M.M.
      • Buntzen S.
      • Bek K.M.
      • Laurberg S.
      Complete obstetric anal sphincter tear and risk of long-term fecal incontinence: a cohort study.
      and 5 studies used ambiguous definitions of FI or separate presentations of solid and liquid stool leakage, which may add up to an overestimated rate of FI.
      • Faltin D.L.
      • Otero M.
      • Petignat P.
      • et al.
      Women’s health 18 years after rupture of the anal sphincter during childbirth: I. Fecal incontinence.
      • Baud D.
      • Meyer S.
      • Vial Y.
      • Hohlfeld P.
      • Achtari C.
      Pelvic floor dysfunction 6 years post-anal sphincter tear at the time of vaginal delivery.
      • Wegnelius G.
      • Hammarström M.
      Complete rupture of anal sphincter in primiparas: long-term effects and subsequent delivery.
      • Evers E.C.
      • Blomquist J.L.
      • McDermott K.C.
      • Handa V.L.
      Obstetrical anal sphincter laceration and anal incontinence 5-10 years after childbirth.
      • Soerensen M.M.
      • Buntzen S.
      • Bek K.M.
      • Laurberg S.
      Complete obstetric anal sphincter tear and risk of long-term fecal incontinence: a cohort study.
      • Halle T.K.
      • Salvesen K.Å.
      • Volløyhaug I.
      Obstetric anal sphincter injury and incontinence 15-23 years after vaginal delivery.
      The prevalence of FI in women without OASI varied from 5.3%
      • Faltin D.L.
      • Otero M.
      • Petignat P.
      • et al.
      Women’s health 18 years after rupture of the anal sphincter during childbirth: I. Fecal incontinence.
      to 12.2%
      • Baud D.
      • Meyer S.
      • Vial Y.
      • Hohlfeld P.
      • Achtari C.
      Pelvic floor dysfunction 6 years post-anal sphincter tear at the time of vaginal delivery.
      and after 1 OASI from 13.5%
      • Faltin D.L.
      • Otero M.
      • Petignat P.
      • et al.
      Women’s health 18 years after rupture of the anal sphincter during childbirth: I. Fecal incontinence.
      to 28.8%.
      • Halle T.K.
      • Salvesen K.Å.
      • Volløyhaug I.
      Obstetric anal sphincter injury and incontinence 15-23 years after vaginal delivery.
      Interestingly, in 4 studies, the mean increase of FI from no OASI to 1 OASI was 13.5%.
      • Baud D.
      • Meyer S.
      • Vial Y.
      • Hohlfeld P.
      • Achtari C.
      Pelvic floor dysfunction 6 years post-anal sphincter tear at the time of vaginal delivery.
      ,
      • Wegnelius G.
      • Hammarström M.
      Complete rupture of anal sphincter in primiparas: long-term effects and subsequent delivery.
      • Evers E.C.
      • Blomquist J.L.
      • McDermott K.C.
      • Handa V.L.
      Obstetrical anal sphincter laceration and anal incontinence 5-10 years after childbirth.
      • Soerensen M.M.
      • Buntzen S.
      • Bek K.M.
      • Laurberg S.
      Complete obstetric anal sphincter tear and risk of long-term fecal incontinence: a cohort study.
      • Halle T.K.
      • Salvesen K.Å.
      • Volløyhaug I.
      Obstetric anal sphincter injury and incontinence 15-23 years after vaginal delivery.
      There are 2 large projects on sphincter injuries, 1 each from Denmark and Sweden, about FI after no OASI, 1 OASI, and 2 OASIs, which were based on validated national registers. The follow-up times were 12 and 20 years, respectively.
      • Jangö H.
      • Langhoff-Roos J.
      • Rosthøj S.
      • Sakse A.
      Recurrent obstetric anal sphincter injury and the risk of long-term anal incontinence.
      ,
      • Nilsson I.
      • Åkervall S.
      • Milsom I.
      • Gyhagen M.
      Long-term effects of vacuum extraction on pelvic floor function: a cohort study in primipara.
      In women without OASI, the prevalence of FI was 5.6% in the Danish study
      • Persson L.K.G.
      • Sakse A.
      • Langhoff-Roos J.
      • Jangö H.
      Anal incontinence after two vaginal deliveries without obstetric anal sphincter rupture.
      and 13.7% in the Swedish study.
      • Nilsson I.
      • Åkervall S.
      • Milsom I.
      • Gyhagen M.
      Long-term effects of vacuum extraction on pelvic floor function: a cohort study in primipara.
      After 1 OASI, the prevalence rates were 13.2% and 16.5% (Denmark, 2 vaginal deliveries)
      • Jangö H.
      • Langhoff-Roos J.
      • Rosthøj S.
      • Sakse A.
      Recurrent obstetric anal sphincter injury and the risk of long-term anal incontinence.
      ,
      • Jangö H.
      • Langhoff-Roos J.
      • Rosthøj S.
      • Sakse A.
      Mode of delivery after obstetric anal sphincter injury and the risk of long-term anal incontinence.
      and 28.4% (Sweden, 1 vaginal delivery),
      • Nilsson I.
      • Åkervall S.
      • Milsom I.
      • Gyhagen M.
      Long-term effects of vacuum extraction on pelvic floor function: a cohort study in primipara.
      and after 2 OASIs, the prevalence rates were 23.6%
      • Jangö H.
      • Langhoff-Roos J.
      • Rosthøj S.
      • Sakse A.
      Recurrent obstetric anal sphincter injury and the risk of long-term anal incontinence.
      and 36.1% (both 2 pregnancies), respectively. In all, the difference in prevalence among the national cohorts was more or less 10% after no OASI, 1 OASI, and 2 OASIs but overall 10% higher in the Swedish groups than in the Danish groups. This might be explained by the difference in follow-up time and women’s age at the time of reporting. In support of this assumption, a longitudinal study from the Netherlands reported that the prevalence of FI (once a week or more often) increased from 8.8% (11/125) at a follow-up of 14 years to 15.1% (18/119) at 25 years, which corresponds to an increase of 6.3% during the additional 11 years after the OASI.
      • De Leeuw J.W.
      • Vierhout M.E.
      • Struijk P.C.
      • Hop W.C.
      • Wallenburg H.C.
      Anal sphincter damage after vaginal delivery: functional outcome and risk factors for fecal incontinence.
      ,
      • Mous M.
      • Muller S.A.
      • de Leeuw J.W.
      Long-term effects of anal sphincter rupture during vaginal delivery: faecal incontinence and sexual complaints.

       Clinical implications

      The long-term consequences after 1 OASI and more so after 2 OASIs are appallingly bleak, and the risk of a repeat OASI is distressingly high in women with OASI at the first birth. This information must be communicated to women as part of the antenatal consultation. Avoiding sphincter injuries would most probably significantly reduce the severity, postpone the onset, and even prevent the occurrence of FI.

       Research implications

      Further research should investigate the severity and impact of FI in women with 1 or 2 OASIs and the risk for co-occurring PFDs. Appropriate reference groups of nulliparous women and women who underwent cesarean delivery should be studied to distinguish the effects of the natural course and the consequences of pregnancy in itself from that of sphincter injuries. Large register-based datasets should be used to develop clinically useful preventive strategies to reduce the number of OASIs.

       Strengths

      The main strengths of this study were the Swedish national registers and the study design. The TPR, which is administered by Statistics Sweden, is a fundamental part of epidemiologic research in Sweden. The quality of the TPR and MBR is regarded as high because of their completeness. Statistics Sweden serves medical researchers with data by linking TPR data to other medical registers, using the mandatory 12 digits personal identity number.
      • Ludvigsson J.F.
      • Otterblad-Olausson P.
      • Pettersson B.U.
      • Ekbom A.
      The Swedish personal identity number: possibilities and pitfalls in healthcare and medical research.
      The study design used restriction to control confounding.
      • Rothman K.J.
      • Greenland S.
      • Lash T.L.
      Design strategies to improve study accuracy, restriction.
      The study cohorts were large, although the study was restricted to women who had 2 pregnancies registered in Sweden, with 2 singleton births, and no further deliveries, with the exclusion of multifetal and ongoing pregnancies. Information was retrieved from a homogeneous Nordic population and within a limited period concerning childbirth and data collection. We used the large cohort of women who had 2 vaginal deliveries without OASI as a reference for control of intrinsic and external factors not related to sphincter injury. This cohort (also including women with 1 OASI) was randomly selected from a national source cohort. The response rates (66% and 70%) in this study were high compared with those in most current studies in this field of research.
      • Handa V.L.
      • Blomquist J.L.
      • Knoepp L.R.
      • Hoskey K.A.
      • McDermott K.C.
      • Muñoz A.
      Pelvic floor disorders 5-10 years after vaginal or cesarean childbirth.
      ,
      • Dolan L.M.
      • Hilton P.
      Obstetric risk factors and pelvic floor dysfunction 20 years after first delivery.
      However, the predisposition to participate in questionnaire studies is higher in symptomatic women, whereby the prevalence of PFDs might be overestimated.
      • Thom D.H.
      • van den Eeden S.K.
      • Ragins A.I.
      • et al.
      Differences in prevalence of urinary incontinence by race/ethnicity.

       Limitations

      The diagnosis of FI was not based on any objective measurement. FI is, however, a symptom, and this perception is the source on which any evaluation and impact of incontinence rest and can affect the willingness and ability of women to perceive, evaluate, and report correctly. However, to survey large populations, a questionnaire study is considered the most suitable tool for gathering information about sensitive issues, which especially applies to all matters involving FI.
      • Hartge P.
      • Cahill J.
      Field methods in epidemiology.
      Furthermore, several studies have indicated that self-reporting is consistent
      • Grodstein F.
      • Fretts R.
      • Lifford K.
      • Resnick N.
      • Curhan G.
      Association of age, race, and obstetric history with urinary symptoms among women in the Nurses’ Health Study.
      and valid when the symptom exists at the time of reporting.
      • Diokno A.C.
      • Brown M.B.
      • Brock B.M.
      • Herzog A.R.
      • Normolle D.P.
      Clinical and cystometric characteristics of continent and incontinent noninstitutionalized elderly.
      ,
      • Herzog A.R.
      • Fultz N.H.
      Prevalence and incidence of urinary incontinence in community-dwelling populations.
      A second weakness was the reliability of data about OASI in MBR during the study period. Underreporting of OASI, which was recognized after the advent of endoanal ultrasonography in 1989,
      • Law P.J.
      • Bartram C.I.
      Anal endosonography: technique and normal anatomy.
      is widely recognized as a considerable obstacle in epidemiologic research of OASI and the etiology of FI.
      • Sultan A.H.
      • Kamm M.A.
      • Hudson C.N.
      • Thomas J.M.
      • Bartram C.I.
      Anal-sphincter disruption during vaginal delivery.
      ,
      • Andrews V.
      • Sultan A.H.
      • Thakar R.
      • Jones P.W.
      Occult anal sphincter injuries--myth or reality?.
      Most probably, tears that involved only part of the sphincter muscle were unreported, which might have led to a selection bias of more severe injuries in the OASI cohorts and hence to overestimation of FI. Conversely, it probably also means a corresponding presence of OASIs of lower degrees in women who were allocated to the no OASI cohort, therefor acting the other way. Furthermore, the original ambition to determine the long-term effect of OASI was not fully realized, as the number of women aged >55 years with 1 or 2 OASIs turned out to be fewer than expected from the outset but crucial to fully capture the increase in the incidence of FI that seemed to occur around menopause.

       Conclusions

      The findings of this report support the assumption that there is a causal link between sphincter injury and long-term FI. The cumulative effect of a repeat OASI indicates a dose-response relationship between OASI and FI. Overall, our results comply with the findings from other fields of research on the pelvic floor that have revealed fractured anatomy, a decreased contractile strength, and posttraumatic neuropathy following a sphincter tear.

      Appendix

      This appendix has been provided by the authors to give readers additional information about their work.
      Figure thumbnail fx1
      Supplemental Figure 1Annual rate of OASI and cesarean delivery
      A, Incidence rate of women with sphincter injury. B, Incidence rate of women who underwent cesarean delivery. Adapted from the Swedish National Board of Health and Welfare.
      The Swedish National Board of Health and Welfare
      Statistics database for pregnancies, births and newborns. Socialstyrelsen.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      Figure thumbnail fx2
      Supplemental Figure 2The Fisher exact conditional test for 2 proportions
      A, Outcome with high proportion in the population. B, Outcome with low proportion in the population.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      Figure thumbnail fx3
      Supplemental Figure 3Follow-up time graphs of women with or without sphincter injury
      A, Follow-up time from first birth. B, Time between first and second birth. C, Time between the first and second births in women with and without sphincter injury at first birth.
      OASI, obstetrical anal sphincter injury.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      Supplemental Table 1The incidence of obstetrical anal sphincter injury in Sweden, 1990–2018
      Primiparous womenMultiparous women
      YearNumberPercentageYearNumberPercentage
      199013192.919904820.8
      199114123.219915450.8
      199215483.619926190.9
      199315774.019936181.0
      199417294.619947641.3
      199516604.819957521.4
      199616655.019966911.4
      199716175.219976701.5
      199816215.619986841.6
      199918486.219997581.8
      200021466.720007641.8
      200121696.720018061.9
      200222926.620027871.8
      200324136.820038451.9
      200425257.020048881.9
      200524156.820057311.3
      200623666.420067941.7
      200722596.020077761.6
      200823065.920087461.5
      200921655.520098271.7
      201024886.020108761.7
      201123536.020118341.6
      201223055.820128631.7
      201324556.320138801.7
      201423736.020148311.5
      201521765.420157311.3
      201621005.120167441.3
      201720645.120177051.3
      201818374.520186151.1
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      Supplemental Table 2The incidence of cesarean delivery in Sweden, 1990–2018
      Primiparous womenMultiparous women
      YearNumberPercentageYearNumberPercentage
      1990602711.7199067939.8
      1991629612.4199169719.8
      1992606312.5199270019.7
      1993609513.3199368349.9
      1994593113.7199465609.9
      1995565414.01995605910.1
      1996558714.51996564510.3
      1997554015.21997578011.3
      1998580916.61998583111.8
      1999594216.61999604412.4
      2000669217.32000635612.9
      2001735218.62001689214.0
      2002776918.42002729414.4
      2003800218.52003778614.7
      2004843319.02004821515.0
      2005850319.42005853315.5
      2006902819.72006920016.2
      2007903219.42007921416.1
      2008912518.92008913715.7
      2009940419.32009939115.8
      2010951418.82010951418.8
      2011905918.72011905918.7
      2012935019.02012935019.0
      2013952619.62013952619.6
      2014963919.62014963919.6
      2015954819.22015954819.2
      2016994119.32016994119.3
      2017913918.62017913918.6
      2018928018.62018928018.6
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      Supplemental Table 3Characteristics of responders vs nonresponders
      Adapted from Statistics Sweden’s Technical Report 2015.
      Responders (N=9953 [64.5%])Nonresponders (N=5476 [35.5%])
      VariablesSubgroupn%n%
      Age group
      30–396842.29357.8
      40–4958360.7378029.3
      50–59395271.6157128.4
      60–6910276.13223.9
      Country of birth
      Sweden914265.9472734.1
      Abroad81152.074948.0
      Citizenship
      Swedish980764.6537035.4
      Foreign14657.910642.1
      Marital status
      Married588068.2274831.8
      Unmarried236460.1157239.9
      Divorced162259.2111840.8
      Widowed8769.63830.4
      Income (SEK)
      No income8750.38649.7
      1–124,99931549.232550.8
      125,000–199,00062353.254946.8
      200,000–279,999210160.5137339.5
      280,000–369,999363966.0187834.0
      ≥370,000318871.6126528.4
      Communities
      Large city137664.874835.2
      Suburb to large city197264.9106535.1
      Medium-sized town303765.6159134.4
      Suburb to a medium-sized town32364.218535.8
      Commuting municipality74562.245237.8
      Tourism municipality30464.117035.9
      Goods producing municipality77660.450839.6
      Sparsely populated municipality13466.36833.7
      Densely populated region96166.049434.0
      Sparsely populated region31661.819538.2
      Education
      Unknown650.0650.0
      Compulsory school28942.439257.6
      High school471760.7305239.3
      Upper secondary, <3 y208369.591630.5
      Upper secondary, ≥3 y285872.0111028.0
      Study cohorts
      Randomly selected cohort of women with 2 vaginal deliveries717764.9384235.1
      Randomly selected cohort of women with 2 cesarean deliveries251162.8148937.2
      Total cohort of women with 2 obstetrical anal sphincter injuries32469.114530.9
      SEK, Swedish krona.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      Supplemental Table 4The influence of a sphincter tear at the first delivery on the time to the second birth
      VariablesWomen without OASI at the first delivery (n =6760+104=6864)
      This group consists of all women with no OASI plus those with OASI at the second delivery only


      Mean (SD)
      SD is ±1 SD from the mean


      Median (IQR)
      IQR (Q1–Q3)
      Women with OASI at the first delivery (n=324+253=577)
      This group consists of women with 2 OASIs plus those with OASI at the first delivery only


      Mean (SD)

      Median (IQR)
      Difference between groups

      Mean (95% CI)
      The calculation of the CI was based on bootstrapping of 10,000 replicates, picking the 2.5 and 97.5 percentiles of the 10.000 mean differences as the CI.
      P value
      For comparison between groups, the Mann-Whitney U test was used
      Age at first delivery (y)27.8 (4.0)

      27.5 (25.0–30.3)
      28.7 (3.8)

      28.4 (25.9–31.1)
      0.91 (0.59–1.24)<.0001
      Time between the first and second deliveries (y)2.6 (1.0)

      2.4 (1.9–3.2)
      2.8 (1.2)

      2.6 (2.0–3.3)
      0.20 (0.11–0.30)<.0006
      CI, confidence interval; IQR, interquartile range; OASI, obstetrical anal sphincter injury; SD, standard deviation.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      a This group consists of all women with no OASI plus those with OASI at the second delivery only
      b This group consists of women with 2 OASIs plus those with OASI at the first delivery only
      c For comparison between groups, the Mann-Whitney U test was used
      d SD is ±1 SD from the mean
      e IQR (Q1–Q3)
      f The calculation of the CI was based on bootstrapping of 10,000 replicates, picking the 2.5 and 97.5 percentiles of the 10.000 mean differences as the CI.
      Supplemental Table 5Estimates of the effect of age on different aspects of bowel incontinence from the logistic regression models
      CohortAge period (y)ß0ß1P value for ß1
      FI
      No OASI40–60−4.95660.5929<.0001
      • 1
        OASI
      ≤52−5.3428−0.2800.47
      ≥52−5.34281.0444.0492
      • 2
        OASIs
      ≤52−4.69500.0510.89
      ≥52−4.69500.7305.0606
      Isolated gas incontinence
      No OASI40–60−2.11640.0353<.0001
      • 1
        OASI
      40–60−2.10360.0386.0642
      • 2
        OASIs
      40–600.2770−0.0129.52
      Continence for stool and gas
      No OASI40–602.8594−0.0601<.0001
      • 1
        OASI
      40–602.5020−0.0663.0044
      • 2
        OASIs
      40–600.6566−0.0371.13
      To convert the results from the logistic regression to probabilities of outcomes for bowel incontinence at any specific age between 40 and 60 years, the following formula was used: P(Y)=1/[1+e-(ß01×age)], where P(Y) is the probability of the incontinence variable, ß0 is the intercept, and ß1 is the effect of age/year. For the segmented logistic regression models, 2 formulas were used, 1 for women aged <52 years, P(Y<52)=1/[1+e-(ß0 1×age+ß2 ×52)], and 1for women aged >52 years, P(Y>52)=1/[1+e-(ß01×52+ß2×age)], where P(Y<52 or >52) is the probability of FI in women younger than or older than 52 years, ß0 is the intercept, ß1 is the effect of age <52 years, and ß2 is the effect of age >52 years.
      FI, fecal incontinence; OASI, obstetrical anal sphincter injury.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      Supplemental Table 6The estimated probability of fecal incontinence, isolated gas incontinence, and continence to stool and gas
      Study cohortTime points of age (y)
      4045505560
      Value
      The “value” is the chance (0–100) of having FI, isolated gas incontinence, or being completely continent for stool and gas
      95% CI
      The estimated probability and 95% CI was obtained from 5 independent logistic regression models as indicated in Figure 4, C, and Supplemental Table 5
      Value95% CIValue95% CIValue95% CIValue95% CI
      FI
      No OASI7.05.8–8.49.28.3–10.212.011.2–12.815.514.0–17.119.816.8–23.2
      • 1
        OASI
      26.315.6–40.323.718.8–30.321.217.4–25.625.820.6–32.137.022.7–53.5
      • 2
        OASIs
      33.418.2–52.833.925.9–42.934.529.9–39.339.834.4–45.548.837.1–60.6
      Isolated gas incontinence
      No OASI33.130.4–35.837.135.5–38.841.340.1–42.545.643.5–47.850.046.5–53.6
      • 1
        OASI
      36.426.4–47.041.034.8–47.445.741.6–49.850.543.9–57.255.344.3–65.8
      • 2
        OASIs
      44.132.8–56.742.535.6–49.740.936.8–45.239.434.1–45.037.929.2–47.4
      Continence to stool and gas
      Not having either FI or isolated gas incontinence.
      No OASI61.258.3–64.053.952.1–55.646.445.1–47.639.037.0–41.132.129.1–35.3
      • 1
        OASI
      46.235.0–57.838.232.0–44.730.727.0–34.624.118.9–30.218.612.1–27.6
      • 2
        OASIs
      30.420.1–44.726.720.6–33.723.219.8–27.020.115.9–25.017.311.4–25.3
      CI, confidence interval; OASI, obstetrical anal sphincter injury.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.
      a The “value” is the chance (0–100) of having FI, isolated gas incontinence, or being completely continent for stool and gas
      b The estimated probability and 95% CI was obtained from 5 independent logistic regression models as indicated in Figure 4, C, and Supplemental Table 5
      c Not having either FI or isolated gas incontinence.
      Questionnaire
      Questionnaire
      Q1.How tall are you? (cm)
      Q2.How much do you weigh? (kg)
      Q3.Have you given birth?Yes
      If yes, how many children have you had?_______
      No, I have not given birth
      Q4.Do you still have menstrual periods?Yes or no
      If yes, go to Q6
      Q5.If you have no menstrual periods, what is the cause?
      a.Are you pregnant?Yes or no
      b.Has your uterus been removed?Yes or no
      c.Do you use an intrauterine hormone device?Yes or no
      d.Are you in the menopause stage?Yes or no
      e.Do you use estrogen medications?Yes or no
      f.Other causes?Yes or no
      Symptoms from the urinary tract
      Q6.How many times do you urinate during the daytime, on average?_____
      Q7.Do you have to urinate during the night?Yes
      If yes, how many times?____
      No
      Q8.Do you have urinary urgency with a sudden and strong urge to void, which is hard to postpone?Yes
      No
      If no, go to Q10
      Q9.How does your urinary urgency affect you?No problem
      A minor nuisance
      Some bother
      Much bother
      A major problem
      Q10.Do you take any medication for urinary urgency?Yes or no
      Q11.Were you a bed wetter during childhood (involuntary loss of urine while sleeping)?Yes
      If yes, at what age did it stop?
      No
      Q12.Do you have involuntary loss of urine?Yes or no
      If no, go to Q21
      Q13.How often do you have involuntary loss of urine?Less than once a month
      Once or more per month
      Once or more per week
      Every day and/or night
      Q14.How much urine do you leak each time?A few drops
      Small amounts
      Large amounts
      Q15.Do you have involuntary loss of urine in connection with coughing, sneezing, laughing, or lifting heavy items?Yes or no
      Q16.Do you have involuntary loss of urine in connection with a sudden and strong urge to void?Yes or no
      Q17.For how long have you had involuntary loss of urine?0–5 years
      5–10 years
      More than 10 years
      Q18.Have you consulted a doctor because of involuntary loss of urine?Yes or no
      Q19.How does your urinary leakage affect you?No problem
      A minor nuisance
      Some bother
      Much bother
      A major problem
      Q20.If you have given birth, did you have urinary leakage even before the first pregnancy?Yes or no
      Q21.Have you had any surgery for a urinary incontinence?Yes or no
      Q22.Do you take any medication for urinary incontinence?Yes or no
      Q23.Has your mother suffered from urinary leakage?Yes, no, or do not know
      Symptoms from the vagina
      Q24.Do you have a sensation of tissue protrusion (a vaginal bulge) from your vagina?Often
      Sometimes
      Infrequently
      Never
      Q25.Do you suffer from a chafing or a rubbing feeling in your vagina or vulva?Often
      Sometimes
      Infrequently
      Never
      Q26.Do you have to lift the front vaginal wall to start or complete voiding?Often
      Sometimes
      Infrequently
      Never
      If you have no discomforts from your vagina, proceed to Q29.
      Q27.Are your symptoms worse during straining, for example, during heavy lifting?Unchanged
      Better
      Worse
      Q28.How do these vaginal symptoms affect you?No problem
      A minor nuisance
      Some bother
      Much bother
      A major problem
      Q29.Have you received (any) treatment for a prolapse?Yes or no
      Q30.Have you had any surgery for a prolapse?Yes or no
      Q31.Has your mother suffered from prolapse?Yes, no, or do not know
      Symptoms from your back passage
      Q32.Do you leak solid feces involuntarily?Never
      Less than once a month
      Several times a month but less than once a week
      Once a week or more
      Once a day or more
      Q33.Do you leak liquid feces involuntarily?Never
      Less than once a month
      Several times a month but less than once a week
      Once a week or more
      Once a day or more
      Q34.Do you leak flatus or gas involuntarily?Never
      Less than once a month
      Several times a month but less than once a week
      Once a week or more
      Once a day or more
      Q35.Do you use a protective product or pad because of involuntary leakage from the back passage?Never
      Less than once a month
      Several times a month but less than once a week
      Once a week or more
      Once a day or more
      Q36.Is your daily life style affected by involuntary leakage from your back passage?Never
      Less than once a month
      Several times a month but less than once a week
      Once a week or more
      Once a day or more
      Q37.How do your bowel symptoms affect you?No problem
      A minor nuisance
      Some bother
      Much bother
      A major problem
      Q38.Have you received (any) treatment for leakage of flatus or gas or feces?Yes or no
      Q39.Has your mother suffered from leakage of flatus or gas or feces?Yes, no, or do not know
      Q40.On the lines below, there is room for your own comments regarding this questionnaire.
      Nilsson et al. Symptoms of fecal incontinence two decades after no, one, or two obstetrical anal sphincter injuries. Am J Obstet Gynecol 2021.

      References

      1. Sultan A. Thakar R. Fenner D. Perineal and Anal Sphincter Trauma. Springer-Verlag Ltd, London, England2009
        • Bols E.M.
        • Hendriks E.J.
        • Berghmans B.C.
        • Baeten C.G.
        • Nijhuis J.G.
        • De Bie R.A.
        A systematic review of etiological factors for postpartum fecal incontinence.
        Acta Obstet Gynecol Scand. 2010; 89: 302-314
        • Kirss J.
        • Pinta T.
        • Böckelman C.
        • Victorzon M.
        Factors predicting a failed primary repair of obstetric anal sphincter injury.
        Acta Obstet Gynecol Scand. 2016; 95: 1063-1069
        • Webb S.S.
        • Hemming K.
        • Khalfaoui M.Y.
        • et al.
        An obstetric sphincter injury risk identification system (OSIRIS): is this a clinically useful tool?.
        Int Urogynecol J. 2017; 28: 367-374
        • Waldenström U.
        • Ekéus C.
        Risk of obstetric anal sphincter injury increases with maternal age irrespective of parity: a population-based register study.
        BMC Pregnancy Childbirth. 2017; 17: 306
        • The Swedish National Board of Health and Welfare
        Statistics database for pregnancies, births and newborns. Socialstyrelsen.
        (Available at:) (Accessed Sept. 10, 2020)
        • Ekéus C.
        • Nilsson E.
        • Gottvall K.
        Increasing incidence of anal sphincter tears among primiparas in Sweden: a population-based register study.
        Acta Obstet Gynecol Scand. 2008; 87: 564-573
        • Laine K.
        • Gissler M.
        • Pirhonen J.
        Changing incidence of anal sphincter tears in four Nordic countries through the last decades.
        Eur J Obstet Gynecol Reprod Biol. 2009; 146: 71-75
        • Gurol-Urganci I.
        • Cromwell D.A.
        • Edozien L.C.
        • 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: 1516-1525
        • Sultan A.H.
        • Kamm M.A.
        • Hudson C.N.
        • Thomas J.M.
        • Bartram C.I.
        Anal-sphincter disruption during vaginal delivery.
        N Engl J Med. 1993; 329: 1905-1911
        • Sultan A.H.
        Editorial: obstetric perineal injury and anal incontinence.
        Clin Risk. 1999; 5: 193-196
        • LaCross A.
        • Groff M.
        • Smaldone A.
        Obstetric anal sphincter injury and anal incontinence following vaginal birth: a systematic review and meta-analysis.
        J Midwifery Womens Health. 2015; 60: 37-47
        • Olsson F.
        • Berterö C.
        Living with faecal incontinence: trying to control the daily life that is out of control.
        J Clin Nurs. 2015; 24: 141-150
        • Rasmussen J.L.
        • Ringsberg K.C.
        Being involved in an everlasting fight--a life with postnatal faecal incontinence. A qualitative study.
        Scand J Caring Sci. 2010; 24: 108-115
        • Johanson J.F.
        • Lafferty J.
        Epidemiology of fecal incontinence: the silent affliction.
        Am J Gastroenterol. 1996; 91: 33-36
        • Meyer I.
        • Richter H.E.
        Impact of fecal incontinence and its treatment on quality of life in women.
        Womens Health (Lond). 2015; 11: 225-238
        • Bharucha A.E.
        • Zinsmeister A.R.
        • Locke G.R.
        • et al.
        Prevalence and burden of fecal incontinence: a population-based study in women.
        Gastroenterology. 2005; 129: 42-49
        • Sandvik H.
        • Hunskaar S.
        • Seim A.
        • Hermstad R.
        • Vanvik A.
        • Bratt H.
        Validation of a severity index in female urinary incontinence and its implementation in an epidemiological survey.
        J Epidemiol Community Health. 1993; 47: 497-499
        • Tegerstedt G.
        • Miedel A.
        • Maehle-Schmidt M.
        • Nyren O.
        • Hammarström M.
        A short-form questionnaire identified genital organ prolapse.
        J Clin Epidemiol. 2005; 58: 41-46
        • Jorge J.M.
        • Wexner S.D.
        Etiology and management of fecal incontinence.
        Dis Colon Rectum. 1993; 36: 77-97
        • Haylen B.T.
        • De Ridder D.
        • Freeman R.M.
        • et al.
        An International Urogynecological Association (IUGA)/International Continence Society (ICS) joint report on the terminology for female pelvic floor dysfunction.
        Neurourol Urodyn. 2010; 29: 4-20
        • Jha S.
        • Parker V.
        Risk factors for recurrent obstetric anal sphincter injury (rOASI): a systematic review and meta-analysis.
        Int Urogynecol J. 2016; 27: 849-857
        • Faltin D.L.
        • Otero M.
        • Petignat P.
        • et al.
        Women’s health 18 years after rupture of the anal sphincter during childbirth: I. Fecal incontinence.
        Am J Obstet Gynecol. 2006; 194: 1255-1259
        • Baud D.
        • Meyer S.
        • Vial Y.
        • Hohlfeld P.
        • Achtari C.
        Pelvic floor dysfunction 6 years post-anal sphincter tear at the time of vaginal delivery.
        Int Urogynecol J. 2011; 22: 1127-1134
        • Wegnelius G.
        • Hammarström M.
        Complete rupture of anal sphincter in primiparas: long-term effects and subsequent delivery.
        Acta Obstet Gynecol Scand. 2011; 90: 258-263
        • Evers E.C.
        • Blomquist J.L.
        • McDermott K.C.
        • Handa V.L.
        Obstetrical anal sphincter laceration and anal incontinence 5-10 years after childbirth.
        Am J Obstet Gynecol. 2012; 207: 425.e1-425.e6
        • Soerensen M.M.
        • Buntzen S.
        • Bek K.M.
        • Laurberg S.
        Complete obstetric anal sphincter tear and risk of long-term fecal incontinence: a cohort study.
        Dis Colon Rectum. 2013; 56: 992-1001
        • Halle T.K.
        • Salvesen K.Å.
        • Volløyhaug I.
        Obstetric anal sphincter injury and incontinence 15-23 years after vaginal delivery.
        Acta Obstet Gynecol Scand. 2016; 95: 941-947
        • Jangö H.
        • Langhoff-Roos J.
        • Rosthøj S.
        • Sakse A.
        Recurrent obstetric anal sphincter injury and the risk of long-term anal incontinence.
        Am J Obstet Gynecol. 2017; 216: 610.e1-610.e8
        • Nilsson I.
        • Åkervall S.
        • Milsom I.
        • Gyhagen M.
        Long-term effects of vacuum extraction on pelvic floor function: a cohort study in primipara.
        Int Urogynecol J. 2016; 27: 1051-1056
        • Persson L.K.G.
        • Sakse A.
        • Langhoff-Roos J.
        • Jangö H.
        Anal incontinence after two vaginal deliveries without obstetric anal sphincter rupture.
        Arch Gynecol Obstet. 2017; 295: 1399-1406
        • Jangö H.
        • Langhoff-Roos J.
        • Rosthøj S.
        • Sakse A.
        Mode of delivery after obstetric anal sphincter injury and the risk of long-term anal incontinence.
        Am J Obstet Gynecol. 2016; 214: 733.e1-733.e13
        • De Leeuw J.W.
        • Vierhout M.E.
        • Struijk P.C.
        • Hop W.C.
        • Wallenburg H.C.
        Anal sphincter damage after vaginal delivery: functional outcome and risk factors for fecal incontinence.
        Acta Obstet Gynecol Scand. 2001; 80: 830-834
        • Mous M.
        • Muller S.A.
        • de Leeuw J.W.
        Long-term effects of anal sphincter rupture during vaginal delivery: faecal incontinence and sexual complaints.
        BJOG. 2008; 115: 234-238
        • Ludvigsson J.F.
        • Otterblad-Olausson P.
        • Pettersson B.U.
        • Ekbom A.
        The Swedish personal identity number: possibilities and pitfalls in healthcare and medical research.
        Eur J Epidemiol. 2009; 24: 659-667
        • Rothman K.J.
        • Greenland S.
        • Lash T.L.
        Design strategies to improve study accuracy, restriction.
        in: Rothman K.J. Greenland S. Lash T.L. Modern epidemiology. Lippincott Williams & Wilkins, Philadelphia, United States2008. 168–182
        • Handa V.L.
        • Blomquist J.L.
        • Knoepp L.R.
        • Hoskey K.A.
        • McDermott K.C.
        • Muñoz A.
        Pelvic floor disorders 5-10 years after vaginal or cesarean childbirth.
        Obstet Gynecol. 2011; 118: 777-784
        • Dolan L.M.
        • Hilton P.
        Obstetric risk factors and pelvic floor dysfunction 20 years after first delivery.
        Int Urogynecol J. 2010; 21: 535-544
        • Thom D.H.
        • van den Eeden S.K.
        • Ragins A.I.
        • et al.
        Differences in prevalence of urinary incontinence by race/ethnicity.
        J Urol. 2006; 175: 259-264
        • Hartge P.
        • Cahill J.
        Field methods in epidemiology.
        in: Rothman K.J. Greenland S. Lash T.L. Modern epidemiology. Lippincott Williams & Wilkins, Philadelphia, United States2008: 492-510
        • Grodstein F.
        • Fretts R.
        • Lifford K.
        • Resnick N.
        • Curhan G.
        Association of age, race, and obstetric history with urinary symptoms among women in the Nurses’ Health Study.
        Am J Obstet Gynecol. 2003; 189: 428-434
        • Diokno A.C.
        • Brown M.B.
        • Brock B.M.
        • Herzog A.R.
        • Normolle D.P.
        Clinical and cystometric characteristics of continent and incontinent noninstitutionalized elderly.
        J Urol. 1988; 140: 567-571
        • Herzog A.R.
        • Fultz N.H.
        Prevalence and incidence of urinary incontinence in community-dwelling populations.
        J Am Geriatr Soc. 1990; 38: 273-281
        • Law P.J.
        • Bartram C.I.
        Anal endosonography: technique and normal anatomy.
        Gastrointest Radiol. 1989; 14: 349-353
        • Andrews V.
        • Sultan A.H.
        • Thakar R.
        • Jones P.W.
        Occult anal sphincter injuries--myth or reality?.
        BJOG. 2006; 113: 195-200