Obstetric anal sphincter injury (OASI) is a serious complication of vaginal delivery and is associated with long-term anal incontinence (AI) [1]. Sultan introduced his classification over 20 years ago, with the aim of further differentiating OASI according to the degree of sphincter injury (3a - < 50% external anal sphincter, 3b - > 50% external anal sphincter, 3c - internal anal sphincter) [2]. There remains uncertainty regarding the risk of OASI subtypes in the literature. The mechanism of AI may also be influenced by a number of factors (e.g. age, parity), and, therefore, counselling women with OASI is often challenging. The purpose of this systematic review was to address the following questions: (a) are women with major OASI (grade 3c and 4) at higher risk of developing AI when compared to women with minor OASI (grade 3a and 3b)? (b) is a fourth-degree tear more likely to cause AI over a third-degree tear?

A systematic literature search of EMBASE, PubMed, MEDLINE and CINAHL databases from inception until January 2021. We also searched for grey literature, including conference abstracts of International Continence Society (ICS) and International Urogynaecological Association (IUGA) between January 2000-2021. We considered prospective and retrospective cohort studies, cross-sectional and case-control studies without language restrictions. Records were imported into Endnote and deduplicated. Title, abstracts and relevant full text publications were independently screened by two reviewers for inclusion. The quality of cohort and case-control studies was assessed by the Newcastle-Ottawa Scale (NOS), while the Joanna Briggs Institute (JBI) critical appraisal checklist was used for cross-sectional studies. Risk ratios (RRs) were calculated to measure the effect of different grades of OASI. Findings were summarised with narrative synthesis.

The initial search across the electronic databases identified 1847 records, of which 19 full-text articles were included. In addition, 3 conference abstracts were found eligible. Out of 22 studies, 8 each were prospective and retrospective cohort and 6 were cross-sectional studies. Length of follow-up ranged from 1 month to 23 years, with the majority of the reports (n=16) analysing data within 12-months post-partum. We included studies from Europe (n=15), Australia (n=3), US (n=2), Canada (n=1) and Israel (n=1). Overall, 7218 women were included. Third-degree tears evaluated were 6454 versus 764 fourth-degree tears. As per Sultan classification, we identified women (n=2979) with 3a (n=1403), 3b (n=1178) and 3c (n=398) tear. Five studies exclusively included primiparae. Data on minor (3a and 3b) and major (3c and 4th) tears were available in 12 studies. Of these, 3 analyses reported only differences in mean scores in patient-reported outcome measures (PROMs) for AI symptoms.
Women from 8 studies were asked to complete the St. Mark’s Incontinence Score (SMIS), which represented the most popular PROM. This was followed by the Wexner’s score (n=2) and Fecal Incontinence Severity Index (FISI) (n=2). Symptoms were assessed through a non-validated questionnaire in 4 studies. With regard to quality assessment, a medium risk of bias was found in 14 cases, while only 3 were deemed to have a low risk of bias. All studies with high risk of bias (n=5) were retrospective cohort. 
There was a wide variation in the reported prevalence of bowel symptoms and impact of OASI severity across the included studies (Table 1 and 2).  Prospective studies showed that major (3c and 4th) tears are associated with a two-fold risk of AI, when compared to minor (3a and 3b) tears. This was not confirmed by retrospective and cross-sectional studies. On the other hand, retrospective studies consistently showed a risk of faecal incontinence (FI) which was two- to four-fold higher for major tears versus minor tears. Prospective studies showed a trend towards worsening AI symptoms for 4th degree tears but this failed to reach statistical significance. Cross-sectional studies with long-term (≥ 5 years) follow-up showed that women with 4th degree tear were more likely to develop AI (RR 1.41-2.27). Out of 3, 2 retrospective studies showed similar findings but the follow-up was significantly shorter (≤ 1 year). Contrasting results were noted for FI rates, as only 5 out of 10 studies supported an association between 4th degree tear and FI. There were only 2 out of 7 studies which revealed an increased risk of flatal incontinence with 4th degree tears. Interestingly, rates of flatal incontinence were no different when major (3c and 4th) were compared to minor (3a and 3b) tears.

There are few good quality data on the consequences of perineal ruptures. Most studies investigate bowel symptoms within few months from delivery and confounding factors (e.g. age, parity) were often not taken into consideration when drawing conclusions. As such, meta-analysis was not possible due to different assessment tools, length of follow-up and population included. Also, there was no consistency in the definition of bothersome AI as different score thresholds were adopted. Despite the Sultan classification being introduced over 20 years ago, the majority of the data do not provide detailed outcomes for subtypes of third-degree tears. Interestingly, there were only two prospective studies reporting AI rates for women with major tears. Women suffering from 3c tear or above appear to be at increased risk of AI and FI. However, evidence supporting worse outcomes for fourth- versus third-degree tears is less compelling. This may reinforce the importance of the internal anal sphincter in the mechanism of AI [3].

Women are at increased risk of AI following major OASI (grade 3c and 4). There is no clear evidence to support a higher risk of AI for fourth- versus third-degree tears. Well-designed multicentre studies with adequate power and long-term follow-up are needed to evaluate the risk of AI for each OASI subtype.

RCOG Green-top Guideline no.29 (June 2015), London, UK.
Sultan AH. Obstetrical perineal injury and anal incontinence. Clin Risk 1999; 5: 193-196.
Shatari T et al. Importance of evacuatory disturbance in evaluation of faecal incontinence after third degree obstetric tear. Colorectal Disease 2005; 7(1): 18-21.

Continence 2S2 (2022) 100377
DOI: 10.1016/j.cont.2022.100377