Objective
To determine modifiable risk factors and incidence of obstetric anal sphincter injury (OASIS) in primiparous women.
Study Design
We performed a population-based retrospective cohort study, using data from the Danish Medical Birth Registry. The population consisted of primiparous women with a vaginal delivery in the time period 2000-2010. Univariable and multivariable logistic regressions were used to determine risk factors of OASIS. Main outcome measures were incidence of OASIS in first vaginal delivery, odds ratios for possible risk factors: age, body mass index, birthweight, head circumference, gestational age, presentation, induction of labor, oxytocin augmentation, epidural, mediolateral episiotomy, vacuum extraction, forceps, shoulder dystocia, and year of delivery.
Results
Of 214,256 primiparous women with a vaginal delivery, 13,907 (6.5%; 95% confidence interval [CI] 6.4–6.6%) experienced an OASIS. The incidence of OASIS increased in the time period (adjusted odds ratio [aOR], 1.02; 95% CI, 1.02–1.03; P < . 0001, per year). We found a protective effect of epidural analgesia (aOR, 0.84; 95% CI, 0.81–0.88; P = . 0001). Vacuum extraction without episiotomy was a significant risk factor of OASIS (aOR, 2.99; 95% CI, 2.86–3.12; P < . 0001), and episiotomy was protective in vacuum-assisted deliveries compared with vacuum-assisted deliveries without episiotomy (aOR, 0.60; 95% CI, 0.56–0.65; P < . 0001). Birthweight was found to be an important nonmodifiable risk factor (aOR, 2.76; 95% CI, 2.62–2.90; P < . 0001).
Conclusion
Epidural analgesia in itself was protective against OASIS. Vacuum extraction increased the risk of OASIS, although mediolateral episiotomy was protective when applied in deliveries assisted by vacuum extraction.
Obstetric anal sphincter injury (OASIS) is a serious complication to vaginal delivery and is the leading cause of anal incontinence (involuntary loss of flatus, liquid or solid stool) in women. Anal incontinence at long-term follow up has been reported in up to 57% of women with OASIS. The incidence of OASIS in the Nordic countries has been increasing, therefore, it is desirable to identify possible modifiable risk factors of OASIS in order to prevent OASIS and the possible consequences. Previous reports have identified several important risk factors of OASIS such as primiparity, excessive birthweight, vacuum extraction and forceps, whereas mediolateral episiotomy and epidural analgesia has been reported with conflicting effects.
The use of mediolateral episiotomy in the Nordic countries is restricted since liberal use has been reported to cause more perineal trauma, suturing, and healing complications. Median episiotomy has been abandoned because of the increased risk of OASIS. Mediolateral episiotomy has previously been reported as a risk factor, a protective factor, and as an insignificant factor. Some studies have found that mediolateral episiotomy is protective against OASIS if vacuum extraction is used. The effect of epidural has been reported to be protective, insignificant, or even a risk factor in different studies and populations.
The aim of the study was to identify modifiable risk factors and incidence of OASIS in a large population of primiparous women.
Materials and Methods
We performed a population-based cohort study where data were retrieved from the Danish Medical Birth Registry (MBR). Primiparous women with a vaginal singleton delivery at term (fully 37 weeks of gestation) in the time period 2000-2010 (n = 214,256) were included. An OASIS was classified according to the Royal College of Obstetricians and Gynaecologists classification. A third-degree OASIS was defined as a partial or complete disruption of the anal sphincter muscles, which may involve only the external anal sphincter or both the external anal sphincter and the internal anal sphincter. A fourth-degree OASIS was defined as a disruption of the anal sphincter muscles in combination with a tear of the rectal mucosa. OASIS was identified by the International Classification of Diseases 10 codes O70.2 and O70.3, which have been validated in the Danish MBR by comparison with medical records.
We had information on maternal age, maternal prepregnant body mass index (BMI) (reported from 2004), calendar year of delivery, grade of OASIS, birthweight, head circumference, gestational age, presentation, induction of labor, oxytocin augmentation, epidural, mediolateral episiotomy, vacuum extraction, forceps, and shoulder dystocia. All these factors were regarded as potential risk factors of OASIS. We considered induction of labor, oxytocin augmentation, epidural, mediolateral episiotomy, vacuum extraction, and forceps as modifiable factors.
Univariable and multivariable logistic regression analyses were performed to determine the association between risk factors and OASIS. In the multivariable logistic regression analysis, BMI was excluded since it was not significant and a large proportion of the values were missing (40.9%). Odds ratios (ORs) and the corresponding 95% confidence intervals (95% CIs) were calculated. The continuous risk factors were included linearly. We investigated the interaction between episiotomy and vacuum extraction as well as between calendar year and the possible modifiable risk factors. Tests and P values were based on the maximum likelihood principle. P values less than .05 were considered significant. Statistical analyses were performed using SAS version 9.2 (SAS Institute, Cary, NC).
The study was approved by the Danish Data Protection Agency (no. 2012-41-0362).
Results
Of 214,256 women with a first vaginal delivery in 2000-2010, 13,907 (6.5%; 95% CI, 6.4–6.6%) had an OASIS. Of these, 11.5% (n = 1599) had a fourth degree OASIS. Table 1 shows distribution of the potential risk factors in women with and without OASIS. Univariable analyses ( Table 2 ) showed that all factors included in the analysis were significant risk factors, except breech presentation, which was a significant protective factor.
| Potential risk factors | Anal sphincter injury, n = 13,907 | No anal sphincter injury, n = 200,349 |
|---|---|---|
| Maternal factors | ||
| Maternal age, y; median (IQR) | 28 (26–31) | 28 (25–31) |
| Prepregnant BMI, kg/m 2 a ; median (IQR) | 22.8 (20.8–25.7) | 22.6 (20.6–25.4) |
| Calendar year of delivery, median (IQR) | 2005 (2002–2008) | 2005 (2002–2008) |
| Fetal factors | ||
| Birthweight, g b ; median (IQR) | 3700 (3400–4000) | 3460 (3170–3760) |
| Head circumference, cm c ; median (IQR) | 35 (34–36) | 35 (34–36) |
| Gestational age, d; median (IQR) | 283 (278–289) | 282 (275–287) |
| Presentation d | ||
| Occiput anterior | 12,667 (91.5%) | 188,636 (94.5%) |
| Occiput posterior | 667 (4.8%) | 5583 (2.8%) |
| Breech presentation | 17 (0.1%) | 744 (0.4%) |
| Other | 498 (3.4%) | 4620 (2.3%) |
| Obstetric factors | ||
| Induction of labor | 1998 (14.4%) | 25,107 (12.5%) |
| Oxytocin augmentation e | 5599 (40.3%) | 60,691 (30.3%) |
| Epidural | 3356 (24.1%) | 44,193 (22.1%) |
| Episiotomy | 2281 (16.4%) | 30,427 (15.2%) |
| Vacuum extraction | 5323 (38.3%) | 33,665 (16.8%) |
| Forceps | 43 (0.3%) | 238 (0.1%) |
| Shoulder dystocia | 285 (2.0%) | 1310 (0.6%) |
a BMI was reported in 126,732 women
b Birthweight was reported in 213,501 infants
c Head circumference was reported in 210,048 infants
d Presentation was reported in 213,432 deliveries
| Potential risk factors | Univariable analysis n = 214,256 | Multivariable analysis n = 209,687 | |||||
|---|---|---|---|---|---|---|---|
| OR | 95% CI | P value | aOR | 95% CI | P value | ||
| Maternal factors | |||||||
| Maternal age, y | 1.03 | 1.03–1.04 | < .0001 | 1.02 | 1.02–1.03 | < .0001 | |
| Prepregnant BMI a | 1.01 | 1.01–1.02 | < .0001 | — | — | — | |
| Calendar year of delivery | 1.02 | 1.02–1.03 | < .0001 | 1.02 | 1.02–1.03 | < .0001 | |
| Fetal factors | |||||||
| Birthweight, kg b | 3.02 | 2.91–3.14 | < .0001 | 2.76 | 2.62–2.90 | < .0001 | |
| Head circumference, cm c | 1.25 | 1.23–1.26 | < .0001 | 1.02 | 1.01–1.04 | .0035 | |
| Gestational age, d | 1.03 | 1.02–1.03 | < .0001 | 1.00 | 1.00–1.00 | .1014 | |
| Presentation d | |||||||
| Occiput anterior | 1.0 | 1.0 | |||||
| Occiput posterior | 1.78 | 1.64–1.93 | < .0001 | 1.34 | 1.22–1.46 | < .0001 | |
| Breech presentation | 0.34 | 0.21–0.55 | < .0001 | 0.57 | 0.35–0.93 | .0244 | |
| Other presentations | 1.61 | 1.46–1.76 | < .0001 | 1.19 | 1.08–1.32 | .0005 | |
| Obstetric factors | |||||||
| Induction of labor | 1.17 | 1.11–1.23 | < .0001 | 1.04 | 0.99–1.08 | .1261 | |
| Oxytocin augmentation | 1.55 | 1.50–1.61 | < .0001 | 1.14 | 1.10–1.19 | < .0001 | |
| Epidural | 1.12 | 1.08–1.17 | < .0001 | 0.84 | 0.81–0.88 | .0001 | |
| Vacuum extraction | Episiotomy | ||||||
| “no” | “no” | 1.0 | 1.0 | ||||
| “yes” | “no” | 3.42 | 3.29–3.56 | < .0001 | 2.99 | 2.86–3.12 | < .0001 |
| “no” | “yes” | 1.05 | 0.99–1.12 | .1184 | 0.95 | 0.89–1.02 | .1541 |
| “yes” | “yes” | 2.31 | 2.16–2.46 | < .0001 | 1.80 | 1.68–1.93 | < .0001 |
| Forceps | 2.61 | 1.88–3.61 | < .0001 | 1.95 | 1.39–2.75 | .0007 | |
| Shoulder dystocia | 3.18 | 2.79–3.62 | < .0001 | 1.33 | 1.16–1.53 | < .0001 | |
a BMI was reported in 126,732 women
b Birthweight was reported in 213,501 infants
c Head circumference was reported in 210,048 infants
We found an interaction between vacuum extraction and mediolateral episiotomy ( P < .0001). In 18.2% (n = 38,988) the delivery was assisted by vacuum extraction. In vacuum-assisted deliveries, 28.7% (n = 11.178) had a mediolateral episiotomy. If vacuum extraction was used without episiotomy (n = 27,810), 14.9 % had an OASIS (n = 4143; 95% CI, 14.5–15.3%), whereas the risk of OASIS if vacuum extraction was used in combination with mediolateral episiotomy was 10.6% (n = 1180; 95% CI, 10.0–11.1%). The interaction between vacuum extraction and mediolateral episiotomy shows, that when vacuum extraction is used, mediolateral episiotomy is protective against OASIS (adjusted odds ratio [aOR], 0.60; 95% CI, 0.56–0.65; P < .0001), whereas mediolateral episiotomy shows no significant effect on the risk of OASIS in deliveries without vacuum extraction (aOR, 0.95; 95% CI, 0.89–1.02; P = .1520). Numbers needed to treat (NNT) was found to be 23, ie, 23 mediolateral episiotomies has to be performed to prevent 1 case of OASIS in vacuum-assisted deliveries (95% CI, 19.8–27.5).
Epidural was a significant risk factor in the univariable analysis, but became a significant protective factor in the multivariable analysis. When we adjusted for birthweight only, epidural was still found to be a risk factor, and when we adjusted for vacuum extraction only, the effect became nonsignificant. When adjusting for both birthweight and vacuum extraction, epidural became a protective factor of OASIS (OR, 0.94; 95% CI, 0.90–0.98; P = .0028).
The Figure shows that the incidence of OASIS increased in the time period (aOR, 1.02; 95% CI, 1.02–1.03; P < .0001, per year). Frequency of mediolateral episiotomy decreased and the incidence of vacuum extraction remained stable over time. The epidural frequency increased markedly from 2000 to 2010. We found no evidence of interaction between the possible modifiable risk factors (epidural analgesia, forceps, oxytocin augmentation, episiotomy, and vacuum extraction) and calendar year.
In patients with OASIS, 57.7% (95% CI, 56.8–58.5) had one or more modifiable risk factor (vacuum extraction, forceps, or oxytocin augmentation). Of all women with 1 or more modifiable risk factor of OASIS (vacuum extraction, forceps, oxytocin augmentation), 9.4% (95% CI, 9.2–9.6) had an OASIS.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
